USA – Production manager and front of house sound engineer Brian Hardaswick started hearing buzz about Allen & Heath’s dLive series in early 2020. “I hit up Sound Image and they had them in stock, so I decided to try out mixing with some multi-tracks,” recounted Hardaswick. “I fell in love with it right away.” Unfortunately, the ensuing COVID pandemic prevented Hardaswick from taking the console out on the road. “I ended up going to our rehearsal space twice a week for six to eight hours just to practice mixing by myself on a PA.”
Two years later, as live music made its comeback, Hardaswick finally had the chance to take his dLive system on a headline tour with acclaimed heavy rock band In This Moment. “I was blown away by the console: how true the FX are, how responsive the gate and compression is,” noted Hardaswick. “Every little nuance and every little change I make is audible.” Hardaswick also remarked on dLive’s intuitive workflow. “I just sat down at the controls, and it made sense right out of the gate. It’s amazing that the fader layout is fully customisable with a simple drag-and-drop interface.”
The complete system for In This Moment’s tour includes an S5000 control surface, along with a DM64 MixRack which features 64 mic preamps, 32 line outputs and a powerful 128-channel 96kHz FPGA processor. Hardaswick also has a 128-channel WAVES3 module installed, which he uses for multi-tracking and virtual soundcheck with a computer. “The virtual soundcheck feature is an incredible tool. I love how easy it is to switch back and forth between playback and live mixing, especially at festivals when I don’t have much time to dial things in.”
In This Moment lead vocalist Maria Brink poses a unique challenge for Hardaswick, as her live performances involve use of a headworn microphone and lots of movement through noisy crowds. “The dLive’s DYN8 processor is huge for her, because there’s a lot of noise I need to take out but I don’t want to remove anything important from her actual performance.” The DYN8’s DEEP multi-band compressor and dynamic EQ allow for the plug-ins to only be active in affecting the signal when particular thresholds are hit at specific frequencies. “Her vocals have so much diversity. Sometimes she screams, sometimes she whispers, and I have to keep all that audible. DYN8 allows me to pull certain frequencies during screaming vocals, but also to emphasise frequencies when she’s whispering to make it feel more present and ‘in-your-face’.”
Hardaswick also lauded the built-in parallel compression that is available on every input and mix processing channel on dLive. “I used to have to burn my groups to get a compressor wet/dry mix,” explained Hardaswick. “It’s great that I don’t need a bunch of outboard gear, everything I need is right there.”
This year, In This Moment supported fellow heavy rockers Slipknot on their arena tour. They also performed at the Welcome to Rockville heavy metal festival at Daytona Beach, and are headlining at the Blue Ridge Rock festival in Virginia.
© 1999 – 2022 Entertainment Technology Press Limited News Stories
In This Moment Tours with dLive S5000 – EtNow.com
Mo Amer’s ‘Mo’ is a love letter to his heritage and hometown – The Seattle Times
NEW YORK (AP) — Up-and-coming comedian Mo Amer was opening for Dave Chappelle in 2014 when the more established comedian made a suggestion: Amer should make a short film to introduce his act.
“I started thinking about the short film. I put it together, I was obsessed. And then I wrote it out and then all my friends were like, ‘This is genius, bro. You need to save this for a TV series.’ And I was like, ‘Word? TV series?’ Since then, I’ve just been kind of accumulating all these stories from my mother, my family… my history, my refugee experience in America, the fish-out-of-water element.”
Amer’s true family history serves as the foundation of his new Netflix dramedy, “Mo.” He describes the series as a love letter to his Palestinian heritage and his hometown.
“Although it’s a refugee story, an immigrant Palestinian story, it’s also a love letter to Houston. It’s also like an everyman struggle (story) — people who are working paycheck to paycheck, they’re trying to take care of family, people that are dealing with addiction,” said the 41-year-old. “It has all these layers to it.”
The debut season, comprised of eight half-hour episodes, was created by Amer and Emmy-nominated actor Ramy Youssef, Amer’s good friend and former roommate. “Mo,” along with shows like Youssef’s critically acclaimed “Ramy,” is part of a growing — but still extremely underrepresented — movement in Hollywood bringing Muslim and Middle Eastern stories to the masses.
The series follows Mo Najjar (played by Amer) as his family, including his mother (Farah Bsieso) and brother (Omar Elba) navigate through life in the Houston area in ways many immigrants in America are forced to: seeking employment without citizenship and therefore working odd, under-the-table jobs, as well as maneuvering an often unforgiving health care system and trying assimilate into society.
Similar to the storyline, Amer’s own parents were originally from Palestine before fleeing to Kuwait. Amer was born there, but the Gulf War forced them to leave and eventually settle in Texas. Learning about his family history and their struggles through conversations with his mother, Amer, helped create a blueprint. However, he also learned hard, previously unknown truths, including how his father was tortured in Kuwait.
In the series, flashbacks give context, and some are pivotal to both the show and his real life; episode seven opens with his mother cleverly hiding money as the family prepares to escape Kuwait.
“My mom started just crying when she saw it. She was just like, ‘Wow, I can’t believe you recreated this like that.’ She’s like, ‘How did you even remember the house, the colors, the things on the walls?’” Amer recalled proudly. “I can’t forget that.”
Houston is known for many things: space exploration, its world-renowned medical center and its slowed down hip-hop music known as screwed music; what America’s fourth largest city is not known for is a film and TV industry. However, Amer was adamant about shooting there, aiming to raise the show’s authenticity factor. Despite a couple of major films shot in Houston in the early ’90s it hasn’t been a destination for moviemakers.
Local talent throughout Texas was hired for production, and there are a few notable guest appearances, including rising Houston-based rapper Tobe Nwigwe who plays Najjar’s best friend, and cameos from well-known Texas-flag bearers Paul Wall and Bun B of the legendary southern duo, UGK.
“It just really was shocking to me there’s never been a narrative sitcom, native series filmed out of Houston, which is absurd to me knowing what Houston has to offer,” he explained. “Everything is deliberate from the music choice… to the attire, to the accents, to the feel, the picture, the cinematography. There’s nothing that we didn’t think really about.”
“Mo” has become one of 2022’s buzzworthy and binge-worthy shows, attracting viewers from all backgrounds and a deep respect from many Middle Eastern Americans. But he emphasizes the show has been a long time in the making.
“Netflix drops everything all at the same day and it’s just like, ‘BOOM! Here it is.’ But those episodes are many, many years of really thinking this through,” he said. “You hope it resonates in that way, and it’s such a relief when it does. It’s like, ‘They get it. I didn’t screw this up.’” ___
Follow Associated Press entertainment journalist Gary Gerard Hamilton with his handle @GaryGHamilton on all social media platforms.
The making of Amy Winehouse's Back to Black – “Damn, I wish it did take longer, because I would have more memories” – MusicRadar
MusicRadar is supported by its audience. When you purchase through links on our site, we may earn an affiliate commission. Here’s why you can trust us.
Mark Ronson and more on the creation of a classic track, in their own words
Back to Black, the title track of Amy Winehouse’s second and tragically final studio album was recorded at Chung King Studios and Daptone Records in New York in March 2006, before the finishing touches were applied at Metropolis in London.
The album won numerous awards including Grammies for Record of the Year, Pop Vocal of the Year and Song of the Year (for Rehab), narrowly missing out on Album of the Year. Winehouse took home the Best New Artist of the Year, while Mark Ronson was named Producer of the Year (Non-Classical). To date, Back to Black has sold more than 16 million copies.
I instantly loved her
Speaking to NME (opens in new tab) earlier this year, producer Mark Ronson recalled the to-him all too rapid creation of Back to Black’s backbone, focussing on the title track: “The six songs that we did on Back to Black together, we did in five to seven days.
“I always think, ‘Damn, I wish it did take longer’ because I would have more memories… it was just so quick.”
In a TikTok post (below), Ronson recalls their first meeting and the almost instant birth of the song’s backbone.
“I instantly loved her,” he says. “She played me all this great ’60s music, and I got very inspired, and I came up with this piano part”.
Ronson quickly worked up the part, redolent of ‘60s girl groups, before moving on to crack the core of the classic sound required, in the same session. “I was like, ‘OK, what can I do to make this instantly evocative of the music she likes?’” he recalls in his breakdown of the track for Apple TV’s Watch The Sound (opens in new tab), “and I was like, ‘Oh, I know what, I’ll cake this tambourine in reverb.’”
“I just had basic plugins,” Ronson confirms, adding that “A little bit of spring or plate reverb goes a very long way” in his also-excellent BBC Maestro series (opens in new tab), in which he returns to the original demo project in Pro Tools, showing how he used reverb and distortion to create the dusty piano and drum sounds.
The next morning Winehouse returned to continue work on the song. “She came back the next day,” Ronson told Fader (opens in new tab) in 2015. “She didn’t really get animated about anything, ever – she just had this dead stare – so I thought what I did didn’t work.
“And then she just looked at me and said, ‘I love it. That’s what I want my whole album to sound like.’ So she stayed for another five days, just me and her.”
The next challenge was fully capturing the vintage feel both now agreed would define the album’s sonic approach. Brooklyn’s Daptone Records, up the road from Ronson’s own New York HQ, offered the perfect solution.
Its all-analogue, strictly tape studio (opens in new tab) – complete with Trident Series 65 24–input desk, restored 3M M23 quarter–inch two–track deck, Otari MX5050 quarter–inch tape machine and more – plus killer in-house band The Dap-Kings (opens in new tab), who would back the Ronson produced tracks on the album, specialised in faithfully recreating the soul-drenched Stax and Motown feel required.
“There’s an old–school feel to the studio, but it’s by no means some kind of vintage museum,” Roth told Sound on Sound. “Show me a computer that sounds as good as a tape machine and I’ll use it.”
Before recruiting The Dap-Kings, “We were using every computer trick in the book to make it sound old,” Ronson told The New York Times (opens in new tab). “But it was just so ridiculous.”
Show me a computer that sounds as good as a tape machine and I’ll use it
“That was, like, the most professional thing we’d done to date,” Dap-Kings guitarist Binky Griptite told billboard.com. “We’d just been doing our own thing and on our own time in Gabe’s studio, or were at home smoking weed and working on our own schedule.
“We weren’t used to having another producer being in there, like, ‘Okay, we gotta knock this out by five.’ I mean, we’d had a taste of it, but it was still in the early days.”
We weren’t used to having another producer being in there, like, ‘Okay, we gotta knock this out by five.’
Daptone head honcho and Dap-Kings bass player Gabriel Roth recalled the sessions as “something we spent a handful of afternoons knocking out real quick so we can pay some rent.”
“It was a good, honest piece of work,” he continued, speaking to The New York Times, “but our music is something we’ve been working on together as a family for a dozen years, and it’s a product of a lot of love and creativity.”
Winehouse laid down her vocals before Ronson worked with Daptone head honcho Roth and Dap-Kings drummer and band leader Homer Steinwess to complete the recordings, committing everything to actual tape, in line with the Studio’s strict ‘no Pro Tools’ ethos.
Mix engineer Tom Elmhirst discussed the results with Sound on Sound in 2007: “The bulk of the recordings for the songs that Mark produced on the album were done at the studio of this band called the Dap-Kings, in Brooklyn, New York.
“They recorded drums, piano, guitar and bass all together in one room. The drums were recorded with one microphone, and there’s lots of spill between the instruments, which was great.
“Mark then came to London to record strings, brass and percussion in one of Metropolis’s tracking rooms.”
That was the sneaky part: “I had to do the orchestra behind her back,” Ronson told Fader (opens in new tab). “I said, ‘Amy, The Shangri-Las, the Phil Spector shit that you love, it’s the orchestra that’s doing a lot of that work’. She was like, ‘Nah, I fucking hate it. It’s Disney bullshit’.”
This put arranger Chris Elliot, the man tasked with adding the bullshit, in a tricky situation; speaking to Uncut magazine he said “Mark Ronson, Darcus [Beese, from Island Records] and Tom [Elmhirst] heard the mixes and felt there could be another colour in the tracks.
She had her head down on the desk, and I was kind of terrified. When the song was done, she looked up, beaming, and said, ‘OK, you can leave it in’
“Tom mentioned strings. Amy wasn’t really a fan of the idea of strings. In fact she said very, very bluntly to Mark, ‘I don’t want any fucking strings on the record’.”
Ronson picks up the story (Fader (opens in new tab)): “She didn’t hear it until Love Is a Losing Game (opens in new tab). She had her head down on the desk, and I was kind of terrified. When the song was done, she looked up, beaming, and said, ‘OK, you can leave it in’.
“You know, the production was really a Trojan Horse. It dressed the music up to get it through the wall, but that’s not why the record is what it is. It’s her.”
Next up, the final mix was completed at Metropolis.
Elmhirst: “For me, the mix had to have a contemporary feel to it as well, while Mark wanted to keep the mix sparse and not over-produced. I love Amy’s voice and the vibe of the track, so mixing was a matter of getting the feeling right, with the brass, strings and percussion, but also getting it up to date beat-wise, so that it would work for radio without losing any of its charm.
“Although there’s quite a lot going on in these tracks, they hopefully don’t sound ridiculously complicated. The mix came together quite quickly because I knew exactly what I wanted to do.”
The finished product is a stone-cold classic, a fitting centrepiece to the album that definitively captures a generational talent gone too soon.
Ronson himself still feels the reverberations, as he told NME: “I can sometimes hear ‘Back To Black’ in some restaurant in the background and it does nothing, and then I’ll hear it on another occasion in, like, the lobby of a hotel, and it has a really heavy effect on me.
“I mean, she’s definitely there. I couldn’t tell you like, three days a week or whatever. She’s just there ‘cause I feel like… she kinda put me on the map, so all of my success and everything I’ve had since is somehow linked back to this thing, you know?”
But let’s end on a happier note. Recalling the recording of Valerie, Winehouse’s smash hit Zutons cover, Ronson told Yahoo entertainment (opens in new tab), “The really lovely thing about that session was she had never met the musicians from the Dap-Kings, the people that had played on Back to Black, because we wrote and arranged the songs that she wrote, I arranged the songs while she was in New York, and then she had to go back to London, and then we tracked it with the band.”
Talking to Fader (opens in new tab) again: “We went to Brooklyn so she could finally go to Daptone and meet the guys who played on the record.
“She’s so charming; she was there sitting in the room, just laughing. I remember seeing her walk down the street on the way out, walking on top of the yellow lines. It was one of the last times I saw her carefree and anonymous.
“It was really a lovely day.”
I’m lucky enough to be MusicRadar’s Editor-in-chief while being, by some considerable distance, the least proficient musician on the editorial team. An undeniably ropey but occasionally enthusiastic drummer, I’ve worked on the world’s greatest music making website in one capacity or another since its launch in 2007. I hope you enjoy the site – we do.
Don’t miss the latest deals, news, reviews, features and tutorials
Thank you for signing up to Musicradar. You will receive a verification email shortly.
There was a problem. Please refresh the page and try again.
MusicRadar is part of Future plc, an international media group and leading digital publisher. Visit our corporate site (opens in new tab).
© Future Publishing Limited Quay House, The Ambury, Bath BA1 1UA. All rights reserved. England and Wales company registration number 2008885.
Call for applications: Music performance and business training in Cameroon – Music In Africa
*Your language will be set using a browser cookie.
Don’t have an account? Sign up
Forgot your password?
The Lionn Academy for Arts and Entertainment in Cameroon is inviting applications for the first cohort of its music performance and music business training programme.
The initiative, running under the leadership of musician Mr Leo, is being held in partnership with the Canadian High Commission in Cameroon, Local Youth Corner, Escale Bantoo and DJ Fredy Muks.
The goal of the project is to promote the socioeconomic integration of young people, especially women, IDPs and minorities, in arts and entertainment. It is expected to provide orientation and capacity building for 15 semi-professionals to give them a better understanding of the ecosystem and tools that can favour their chances of being successful and building sustainable careers.
This first cohort, comprising 15 young people – 10 for the music performance track and five for the music business track – will undergo three months of training. Four of the trainees in the music performance track will be selected at the end to undergo a two-week production residency programme under the mentorship of Mr Leo.
While the primary goal of the project is training, the candidates will also be assisted with a basic minimum package to help them kickstart their careers including a grand live showcase.
Programme timeline
Benefits of the programme
Selection Criteria
Note:
Interested candidates can submit their applications here.
More information is available at www.lionnacademy.org.
Sponsored
See all
Disclaimer: Music In Africa provides a platform for musicians and contributors to embed music and videos solely for promotional purposes. If any track or video embedded on this platform violates any copyrights please inform us immediately and we will take it down. Please read our Terms of Use for more.
Exclusive: Studio Bank raises $38 million, creates new holding company – The Business Journals
Studio Bank is ending the year flush with cash.
The homegrown lender closed on a $38 million capital raise, while creating a new holding company, Studio Financial Holdings Inc. (OTCMKTS: STBK), which now owns 100% of the bank, CEO Aaron Dorn told the Business Journal.
The funding includes an $18 million equity raise and $20 million in senior unsecured notes issued through the holding company, which can be reinvested in the bank as equity.
The moves are the latest in Studio’s growth from the first Nashville de novo bank in 14 years to one of the 25 largest banks in the city’s competitive lending market. Studio has grown from $470.2 million in total assets, as of June of 2021, to $769 million in total assets, as of June of this year. Studio shed its de nova status at the end of last year, meaning the bank met financial projections laid out when it launched three years ago and is subject to less intense federal and state oversight.
Dorn said the creation of the new holding company, which was approved by shareholders on May 31, will give the bank flexibility in what kind of capital Studio can raise, such as various debt offerings.
“Holding companies end up creating a source of strength for a bank in capital access,” Dorn said. “At Studio Bank, we are growing very vibrantly and to have this kind of a source of capital strength for us to continue to fuel our growth and expansion, that was a driving force. … From a long-term point of view, this is also a great way for us to set up a structure to be able to consider different corporate development initiatives that will allow us to expand how we serve our clients.”
The new funds will be used to support Studio’s growth, Dorn said, including increasing the type and size of loans the bank offers and expansion into new industries, such as medical and dental practices.
Dorn said Studio intended to raise $10 million to $15 million in equity funding, but that the round was oversubscribed in a matter of two weeks. He said 132 of the 192 investors in the round were local to Middle Tennessee.
“We have such tremendous momentum in our progress as a company and this is really meant to allow us to continue to expand organically the way that we have been all along,” Dorn said. “This new capital is going to allow us to continue to bloom what we have already planted.”
© 2022 American City Business Journals. All rights reserved. Use of and/or registration on any portion of this site constitutes acceptance of our User Agreement (updated January 1, 2021) and Privacy Policy and Cookie Statement (updated July 1, 2022). The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of American City Business Journals.
New Master Chief Petty Officer of the Navy Honea Assumes Office – USNI News – USNI News
Master Chief Petty Officer of the Navy (MCPON) James Honea delivers remarks during the MCPON Change of Office ceremony held at Mahan Hall, United States Naval Academy, on Sept. 8, 2022. US Navy Photo
James Honea became the latest master chief petty officer of the Navy during a change of office ceremony held at the Naval Academy last week.
Honea, who has served in the Navy since 1987, previously served as the senior enlisted sailor at U.S. Indo-Pacific Command. He takes over the MCPON role from Russell Smith, who is retiring after more than 30 years in the sea service. Honea will be the 16th MCPON.
The MCPON works as an advisor on enlisted personnel to the chief of naval operations and the chief of naval personnel. As MCPON, Honea said that his top priorities would be quality of life, warfighting competency and professional and character development, according to a news release from the Navy.
“When I was selected for MCPON, I thought heavily on what my tenure means and what I could bring to you all,” Honea said in the release. “I’m glad I chose to be challenged by this adventure and I’m proud of what it has brought me. I thank you all for accepting the same challenge and being on this journey with me. I have always been proud to be your shipmate, and I’m privileged to be your MCPON.”
Former MCPON Russell Smith on Sept. 9, 2022. US Navy Photo
During Smith’s tenure, which began in August 2018, he focused on improvements for the Navy’s sexual assault prevention and response (SAPR) programs and for access to mental health care. He also started the Laying the Keel program, which provided updated leadership development, and the Sailor of the Year program.
Chief of Naval Operations Adm. Michael Gilday, who presided over the ceremony, spoke about Smith’s commitment to the Navy, especially within the role of MCPON.
“Your tireless efforts to provide sailors with mental, morale, and spiritual support has helped make our fleet more resilient and much more effective,” Gilday said in the release. “Your leadership ensured that our most important resource, our people, are ready to serve and defend the Nation we love. Having conducted more than 200 fleet visits all over the world, your outreach and willingness to be there for Sailors and their families is what truly made your service so remarkable.”
Heather Mongilio is a reporter with USNI News. She has a master’s degree in science journalism and has covered local courts, crime, health, military affairs and the Naval Academy. Follow @hmongilio
Audio Recordings from September 11th – ESPN Southwest Florida
Prepare yourself, these recordings are a lot to handle. Please remember these people and their families, and all the victims, today and every day.
Download the app to LISTEN LIVE wherever you are and connect with us like never before!
Don’t miss on pre-sales, member-only contests and member only events.
MIALive: Women In Music Concert Series returns to Nairobi on 16 September – Music In Africa
*Your language will be set using a browser cookie.
Don’t have an account? Sign up
Forgot your password?
The popular Women In Music Concert Series showcasing Kenyan women artists is set to return after a two-year hiatus at the Kenya National Theatre in Nairobi on 16 September.
The concert – featuring performances by Njerae, Akoth Jumadi, Lisa Oduor-Noah and Fena Gitu – is supported by the Music In Africa Foundation (MIAF) and its Music in Africa Live (MIALive) project.
The Women In Music Concert Series began in 2016 with regular shows held at Goethe-Institute Nairobi. It has so far featured more than 30 musicians, including popular acts like Nikita Kering, Atemi, Chemutai Sage and Mayonde.
The series seeks to promote women artists by providing them with a platform to showcase their original music compositions and performance skills.
“It has been our ultimate goal to put up a high-quality live music production,” Women In Music founder and jazz musician Christine Kamau said. “I am delighted that our efforts have gained recognition from international cultural stakeholders whose partnership has enabled us to put together this event, which I believe adds value to the Kenyan creative industry.”
The concert promises to provide an exciting fusion of acoustic, Afro-fusion, R&B and pop flavours for lovers of live music in the Kenyan capital.
The MIALive project was launched in the wake of the COVID-19 pandemic to support music professionals across the African continent. It is supported by the German Federal Foreign Office, Siemens Stiftung and Goethe-Institut.
Sponsored
See all
Disclaimer: Music In Africa provides a platform for musicians and contributors to embed music and videos solely for promotional purposes. If any track or video embedded on this platform violates any copyrights please inform us immediately and we will take it down. Please read our Terms of Use for more.
New Production Music Library Launch – Radio Ink
SyncWeasel in Orlando, Florida, is a new Production Music Library. Launching with 3,000 music tracks, the company offers clients working across all media, fresh music tracks culled from emerging artists around the world.
“Given the ever-increasing number of broadcast TV, cable, satellite, film, streaming, online and digital media platforms and opportunities that exist today for composers and songwriters by which to get their music heard, we felt now would be a great time to throw our hat into the ring,” said Jay Stollman, Director of Business Development. “There’s a huge market out there for fresh, never-before-heard audio, and we are working in a highly collaborative manner with remarkably gifted musicians and composers from every corner of the globe.”
SyncWeasel’s composers have created music for and with many of the world’s leading entertainment industry brands, including major TV networks and streamers, and production companies. The tracks are pre-cleared, ready-to-license music. The company also offers music supervision services and custom music production, upon request.
More information can be found Here.
Share via:
Daily industry headlines plus breaking news bulletins.

The 2022 Radio Wayne Awards Finalists
When Is the Right Time for Syndicated Programming?
SUBSCRIBE
Rico Blanco toes the line of friendship and love in 'Palibot-libot' – listen – Bandwagon
His larger-than-life performance in Araneta may be over and done, but Rico Blanco gives fans a reason to still be on top of everyone’s minds this week. As announced during his concert last Sunday (11 September), his new single, ‘Palibot-libot,’ has been released.
The award-winning artist’s newest track paints a stirring picture of “someone’s steadfast devotion” despite being in a place of one-sided affection. Inspired by a character in a series Blanco filmed in La Union, ‘Palibot-libot’ sings about the beauty and madness that comes with the uncertainty in love.
Released via Sony Music, the Your Universe act’s newest release features prominent synth-pop elements. This shows the visionary’s mastery of different genres, as he has explored similar music production styles in the past, most notably in his album Galactik Fiestamatik.
Listen to Rico Blanco’s ‘Palibot-libot’ here:
2011 – 2022 Bandwagon Pte Ltd
or sign in with your Bandwagon account
Time Travel Tuesday with Bill Hudson – WJHG
PANAMA CITY, Fla. (WJHG/WECP) – It’s time for another edition of Time Travel Tuesday!
Local historian Bill Hudson stopped by the NewsChannel 7 studio to discuss some footage from Bay County’s history to share with viewers.
Check out what “blast from the past” Bill has in store for us in the video below.
Copyright 2022 WJHG. All rights reserved.
Bonnie Raitt: New LP Out Now U.S. Tour Restarts 9/17 – Rock Cellar Magazine
Home /
It’s a time of triumph for Bonnie Raitt, who released an acclaimed new album, Just Like That … in early May.
The 21st studio album from the revered blues-rock singer/songwriter/guitarist, the record landed atop six different Billboard album charts upon its release, proving that Raitt remains a living legend in every way.
Amid a headlining tour of North America, Raitt has also now revisited her Capitol Records music video catalog in remastered High Definition video, bringing together 13 of her Capitol Records era music videos made between 1989 – 2006 in impressive visual format and housed over at her YouTube channel.
As for that tour, it recently brought Bonnie Raitt and her band through 75 shows across the country, many of them to sold-out crowds. The tour will continue with a new leg that begins on Sept. 17 and runs until Nov. 19, with Staples and Cohn splitting special guest duties throughout the trek.
The dates:
With Special Guest Mavis Staples
17-Sept Oxbow RiverStage Napa, CA – SOLD OUT
18-Sept The Greek Theatre Berkeley, CA|
20-Sept Vina Robles Amphitheatre Paso Robles, CA
22-Sept Santa Barbara Bowl Santa Barbara, CA
24-Sept The Greek Theatre Los Angeles, CA
27-Sept The Rady Shell at Jacobs Park San Diego, CA – SOLD OUT
With Special Guest Marc Cohn
30-Sep ASU Gammage Tempe, AZ – SOLD OUT
1-Oct Tucson Music Hall Tucson, AZ – SOLD OUT
4-Oct Kiva Auditorium Albuquerque, NM
5-Oct Abraham Chavez Theatre El Paso, TX
7-Oct Majestic Theatre San Antonio, TX
8-Oct Texas Trust CU Theatre Grand Prairie, TX
2-Nov ACL Live at The Moody Theater Austin, TX – SOLD OUT
4-Nov Smart Financial Centre at Sugar Land Sugar Land, TX
5-Nov Mahalia Jackson Theater New Orleans, LA
8-Nov Thalia Mara Hall Jackson, MS
9-Nov BJCC Concert Hall Birmingham, AL
11-Nov Johnny Mercer Theatre Savannah, GA
12-Nov St. Augustine Amphitheater St Augustine, FL – SOLD OUT
15-Nov King Center Melbourne, FL – SOLD OUT
16-Nov Au-Rene Theater at Broward Center for the Performing Arts Fort Lauderdale, FL
18-Nov Van Wezel Performing Arts Hall Sarasota, FL – SOLD OUT
19-Nov Ruth Eckerd Hall Clearwater, FL
Your email address will not be published.
← Previous Article
Next Article →
September 13, 2022
September 12, 2022
September 9, 2022
Your source for music news, rock & roll, and celebrities.
10 hours ago
bit.ly
Share on Facebook Share on Twitter Share on Linked In Share by Email
A machine learning technique for identifying DNA enhancer regions utilizing CIS-regulatory element patterns | Scientific Reports – Nature.com
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
Scientific Reports volume 12, Article number: 15183 (2022)
234
2
Metrics details
Enhancers regulate gene expression, by playing a crucial role in the synthesis of RNAs and proteins. They do not directly encode proteins or RNA molecules. In order to control gene expression, it is important to predict enhancers and their potency. Given their distance from the target gene, lack of common motifs, and tissue/cell specificity, enhancer regions are thought to be difficult to predict in DNA sequences. Recently, a number of bioinformatics tools were created to distinguish enhancers from other regulatory components and to pinpoint their advantages. However, because the quality of its prediction method needs to be improved, its practical application value must also be improved. Based on nucleotide composition and statistical moment-based features, the current study suggests a novel method for identifying enhancers and non-enhancers and evaluating their strength. The proposed study outperformed state-of-the-art techniques using fivefold and tenfold cross-validation in terms of accuracy. The accuracy from the current study results in 86.5% and 72.3% in enhancer site and its strength prediction respectively. The results of the suggested methodology point to the potential for more efficient and successful outcomes when statistical moment-based features are used. The current study’s source code is available to the research community at https://github.com/csbioinfopk/enpred.
In cellular biology, regulation of transcription is performed to recruit elongation factors or RNA polymerase II initiation. This is mainly achieved at specific sequences of DNA by binding transcriptional factors (TFs). Transcription initiation sites are harbored by promoter regions which are the most studied sites in DNA1. Some DNA sequences have multiple transcription factor binding sites and are near or far away from promoter regions. Such DNA segments are denoted as enhancers2,3.The transcription of genes is boosted by enhancers which influence various cellular activities such as cell carcinogenesis and virus activity, tissue specificity of gene expression, differentiation and cell growth, regulation and gene expression and develop relationship between such processes very closely4.
Enhancers can be a short (50–1500 bp) segment of DNA and situated 1Mbp (1,000,000 bp) distance away from a gene. Sometimes they can even exist in different chromosomes5,6. On the other hand, promoters are located near the start of the transcription sites of a gene. Due to this fact of locational difference between promoters and enhancers, the task of enhancer’s prediction is highly difficult and challenging than promoters7. Many human diseases like inflammatory bowel disease, disorder and various cancers have been linked to this genetic variation in enhancers8,9,10,11.
A DNA segment characterized as the first enhancer, reported 40 years ago, increased the transcription of β-globin gene during a transgenic assay inside the virus genome of SV40 tumor12. Scientific research during recent past has discovered that enhancers have many subgroups such as weak and strong enhancers, latent enhancers and poised enhancers13. Prediction of enhancers and their subgroups is an interesting area of research as they are considered important in disease and evolution. In higher classification of eukaryotes, transcription factor repertoire, diverse in nature, binds to enhancers14. This process of binding orchestrates many cellular events that are critical to the cellular system. Some of the events that are coordinated through this binding are maintenance of the cell identity, differentiation and response to stimuli15,16.
In the past, purely experimental techniques were being relied upon for the prediction of enhancers. Pioneering works in enhancer prediction was proposed in4,17. The former was to use combinations such as transcription factor, P30018, with enhancers to identify them. This method would usually under-detect or miss the concerned targets. This has resulted in high failure rates because all enhancers do not have transcription factor occupations. The latter was to utilize DNase I hypersensitivity for enhancer predictions. Hence, this led to a high false-positive rate as many other DNA segments, which were non-enhancers, were detected incorrectly as enhancers. Although, genome-wide mapping techniques of histone modifications1,19,20,21,22,23 could improve the aforesaid deficiencies in the prediction of promoters and enhancers, but they are time consuming and expensive.
Several bioinformatics tools have been developed for rapid and cost effective classification of enhancers in genomics. CSI-ANN21 used data transformations efficiently to formulate the samples and predict using Artificial-Neural-Network (ANN) classifications. EnhancerFinder1 incorporated evolutionary conservation information features into sample formulation combined with a multiple kernel learning algorithm as a classifier. RFECS23applied random forest algorithm for improvements in detection methods. EnhancerDBN24 used deep belief networks for enhancer predictions. BiRen25 increased the predictive performance by utilizing deep learning based method. By utilizing these bioinformatics tools, enhancer detection can be achieved by the research community. Formed using many different large sub-groups of functional elements, enhancers can be grouped as weak, strong, inactive and poised enhancers. iEnhancer-2L26,the first ever predictor to detect enhancers and identify their strengths and was based on sequence information only. Pseudo K-tuple nucleotide compositions (PseKNC) based features were incorporated into iEnhancer-2L. It has been used in many analysis related to genomics increasingly. Furthermore, many other methods, such as EnhancerPred27 and EnhancerPred_2.028, were introduced to improve the performance by incorporating other features based on DNA sequences. iEnhancer-5Step29 was recently developed using the hidden information of DNA sequences infused with Support Vector Machine (SVM) based predictions. Recently, iEnhancer-RD30 combined features and utilized recursive feature elimination algorithm for feature selection with deep neural network for enhancer identification. Similarly, ES-ARCNN31 used reverse complement method of data augmentation with residual Convolution Neural Network (CNN) to predict enhancer strength. iEnhancer-GAN32 also implemented CNNs to identify enhancers with strength using deep learning frameworks and combination of word embedding techniques. iEnhancer-XG33 utilized XGBoost classifier as base classifier and five feature extraction methods namely, K-Spectrum Profile, Mismatch K-tuple, Subsequence Profile, Position-Specific-Scoring-Matrix (PSSM) and Pseudo dinucleotide composition (PseDNC) to classify enhancers and their strength. iEnhancer-KL34 also implemented Position specific Nucleotide Composition and Kullback–Leibler (KL) method with several machine learning models. Enhancer-IF utilized comprehensively explored heterogeneous features with five commonly used machine learning algorithms. These five methods were extensively trained using 35 baseline models having seven encodings. This integration of five meta–models enhanced the overall performance of prediction model. BERT(bidirectional encoder representations from transformers)35 and 2D CNN based models were used with the contextualized word embedding for capturing the semantics and context of the words for representing DNA sequences. This opened a new avenue in biological sequence modeling. iEnhancer-MFGBDT36 used gradient boosting decision tree by fusing multiple features which included k-mer, k-mer with reverse compliments, second-order moving components etc. compared to other state of the art methods, this was an effective and intelligent tool to identify enhancers. iEnhancer-ECNN37 used one hot encoding methods and k-mers for data transformation and convolution neural networks (CNN) for identifying enhancers and classify their strengths. An ensemble deep recurrent neural network based method38 was also used to identify enhancers and their strength. These deep ensemble networks were generated from six types of dinucleotide physiochemical properties. These properties outperformed other features and achieved better performance and efficiency. This method proved to be better and has the potential to improve performance of biological sequential modeling using shallow machine learning models. However, improvement in the performance of the aforementioned predictors is still required. Specifically, the success rate of discriminating strong and weak enhancers is not up to the expectations of the scientific community. The current study is initiated to propose a method which would deal with this problem.
The benchmark dataset of DNA enhancer sites, originally constructed and used in recent past by iEnhancer-2L26, was re-used in the proposed method. In the current dataset, information related to nine different cell lines (K562, H1ES, HepG2, GM12878, HSMM, HUVEC, NHEK, NHLF and HMEC) was used in the collection of enhancers and 200 bp fragments were extracted from DNA sequences. The annotation of chromatin state information was performed by ChromHMM. The whole genome profile included multiple histone marks such as, H3K27ac H3K4me1, H3K4me3, etc. To remove pairwise sequences from the dataset, CD-HIT39 tool was used to remove sequences having more than 20% similarity. The benchmark dataset includes 2968 DNA enhancer sequences from which 1484 are non-enhancer sequences and 1484 are enhancer sequences. From 1484 enhancer sequences, 742 are strong enhancers and 742 are weak enhancers for the second layer classification. Furthermore, the independent dataset used by iEnhancer-5Step29 was utilized to enhance the effectiveness and performance of the proposed model. The independent dataset included 400DNA enhancer sequences from which 200 (100 strong and 100 weak enhancers) are enhancers and 200 are non-enhancers. Table 1 includes the breakdown of the benchmark dataset. The details of the above mention dataset is provided in the Supplementary Material (see Online Supporting Information S1, Online Supporting Information S2 and Online Supporting Information S3).
It is not always simple to understand the semantics of a piece of data, which in turn reflects the difficulty of developing biological data models. It can be difficult to come to a consensus about the data in a given domain because different people will emphasize different features, use different terminology, and have different perspectives on how things should be seen. The fact that biosciences are non-axiomatic and that different, though closely related communities have very different perspectives on the same or similar concepts makes the situation even more difficult. Biological data models, however, can be useful for creating, making explicit, and communicating precise and in-depth descriptions of data that is already available or soon to be produced. It is hoped that the current study will increase the use of biological data models in bioinformatics, alleviating the management and sharing issues that are currently becoming more and more problematic.
In statistical based prediction models, the benchmark dataset mostly includes training datasets and testing datasets. By utilizing various benchmark datasets, results obtained are computed from fivefold and tenfold cross-validations. The definition of a benchmark dataset is used in Eq. (1):
where ({D}^{+}) contains 1484 enhancers and ({D}^{-}) contains 1484 non-enhancers. ({D}_{strong}^{+}) contains 742 strong enhancers, ({D}_{weak}^{+}) contains 742 weak enhancers and U denotes the symbol of “union” in the set theory.
An effective bioinformatics predictor is the need of researchers in medicine and pharmacology to formulate the biological sequence with a vector or a discrete model without losing any key-order characteristics or sequence-pattern information. The reason for this fact, as explained in a comprehensive state-of-the-art review40, that the existing machine-learning algorithms cannot handle sequences directly but rather in vector formulations. However, there exists some possibility that all the sequence-pattern information from a vector might be lost in a discrete model formulation. To overcome the sequence-pattern information loss from proteins, Chou proposed pseudo amino acid composition (PseAAC)41. In almost all areas of bioinformatics and computational proteomics40, the Chou’s PseAAC concept has been widely used ever since it was proposed. In the recent past, three publicly accessible and powerful softwares, ‘propy’42, ‘PseAAC-Builder’43 and ‘PseAAC-General’44 were developed and the importance and popularity of Chou’s PseAAC in computational proteomics has increased more ever since. ‘PseAAC-General’ calculates Chou’s general PseAAC45 and the other two software generate Chou’s special PseAAC in various modes46. The Chou’s general PseAAC included not only the feature vectors of all the special modes, but also the feature vectors of higher levels, such as “Gene Ontology” mode45, “Functional Domain” mode45 and “Sequential Evolution” mode or “PSSM” mode45. Using PseAAC successfully for finding solutions to various problems relevant to peptide/protein sequences, encouraged the idea to introduce PseKNC (Pseudo K-tuple Nucleotide Composition)47 for generating different feature vectors for DNA/RNA sequences48,49 which proved very effective and efficient as well. In recent times a useful, efficient and a very powerful webserver called ‘Pse-in-One’50 and its recently updated version ‘Pse-in-One2.0’51 were developed that are able to generate any preferred feature vector of pseudo components for DNA/RNA and protein/peptide sequences.
In this study, we utilized the Kmer52 approach to represent the DNA sequences. According to Kmer, the occurrence frequency of ‘n’ neighboring nucleic acids can be represented from a DNA sequence. Hence, by using the sequential model, a sample of DNA having ‘w’ nucleotides is expressed generally as Eq. (2)
where ({Y}_{1}) is represented as the first nucleotide of the DNA sample S, ({Y}_{2}) as the second nucleotide having the 2nd position of occurrence in DNA sample S and so on so fourth ({Y}_{w}) denotes the last nucleotide of the DNA sample. ‘w’ is the total length of the nucleotides in a DNA sample. The ({Y}_{v}) nucleotide can be any four of the nucleotides which can be represented using the aforementioned discrete model. The nucleotide ({Y}_{v}) can be further described using Eq. (3)
Here (in) is the symbol used to represent the set theory ‘member of’ property and 1 ≤ v ≤ n. The components that are defined by the aforementioned discrete model utilize relevant nucleotides useful features to expedite the extraction methods. These components are further used in statistical moments based feature extraction methods.
Statistical moments are quantitative measures that are used for the study of the concentrations of some key configurations in a collection of data used for pattern recognition related problems53. Several properties of data are described by different orders of moments. Some moments are used to reveal eccentricity and orientation of data while some are used to estimate the data size54,55,56,57,58,59. Several moments have been formed by various mathematicians and statisticians based on famous distribution functions and polynomials60,61,62. These moments were utilized to explicate the current problem63.
The moments that are used in calculations of mean, variance and asymmetry of the probability distribution are known as raw moments. They are neither location-invariant nor scale-invariant. Similar type of information is obtained from the Central moments, but these central moments are calculated using the centroid of the data. The central moments are location-invariant with respect to centroid as they are calculated along the centroid of the data, but still they remain scale-variant. The moments based on Hahn polynomials are known as Hahn moments. These moments are neither location-variant nor scale-invariant64,65,66,67. The fact that these moments are sensitive to biological sequence ordered information amplifies the reason to choose them as they are primarily significant in extracting the obscure features from DNA sequences. These features have been utilized in previous research studies54,59,60,61,68,69,70,71,72,73 and have proved to be more robust and effective in extracting core sequence characteristics. The use of scale-invariant moment has consequently been avoided during the current study. The values quantified from utilizing each method enumerate data on its own measures. Furthermore, the variations in data source characteristics imply variations in the quantified value of moments calculated for arbitrary datasets. In the current study, the 2D version of the aforementioned moments is used and hence the linear structured DNA sequence as expressed by Eq. (2) is transformed into a 2D notation. The DNA sequence, which is 1D, is transformed to a 2D structure using row major scheme through the following Eq. (4):
where the sample sequence length is ‘z’ and the2-dimensional square matrix has ‘(d)’ as its dimension. The ordering obtained from Eq. (4) is used to form matrix M (Eq. 5) having ‘m’ rows and ‘m’ columns.
The transformation from M matrix to square matrix M’ is performed using the mapping function ‘Ʀ’. This function is defined as Eq. (6):
If the population of square matrix M’ is done as row major order then, (i=frac{x}{m}+1) and (j=x mod m).
Any vector or matrix, which represents any pattern, can be used to compute different forms of moments. The values of M’ are used to compute raw moments. The moments of a 2D continuous function (Aleft(j, kright)) to order (j + k) are calculated from Eq. (7):
The raw moments of 2D matrix M, with order (j + k) and up to a degree of 3,are computed using the Eq. (7). The origin of data as the reference point and distant components from the origin are assumed and utilized by the raw moments for computations. The 10 moment features computed up to degree-3 are labeled as ({M}_{00}),({M}_{01}), ({M}_{10}), ({M}_{11}),({M}_{02}), ({M}_{20}, {M}_{12}),({M}_{21}), ({M}_{30}) and ({M}_{03}.)
The centroid of any data is considered as its center of gravity. The centroid is the point in the data where it is uniformly distributed in all directions in the relations of its weighted average74,75. The central moments are also computed up to degree-3, using the centroid of the data as their reference point, from the following Eq. (8):
The degree-3 central moments with ten distinct feature sare labeled as ({mu }_{00}), ({mu }_{01}), ({mu }_{10}), ({mu }_{11}),({mu }_{02}),({mu }_{20}),({mu }_{12}), ({mu }_{21}), ({mu }_{30})&({mu }_{03}.) The centroids (overline{a }) and (overline{b }) are calculated from Eqs. (9) and (10):
The Hahn moments are computed by transforming 1D notations into square matrix notations. This square matrix is valuable for the computations of discrete Hahn moments or orthogonal moments as these moments are of 2D form and require a two-dimensional square matrix as input data. These Hahn moments are orthogonal in nature that implies that they possess reversible properties. Usage of this property enables the reconstruction of the original data using the inverse functions of discrete Hahn moments. This further indicates that the compositional and positional features of a DNA sequence are somehow conserved within the calculated moments. M’ matrix is used as 2D input data for the computations of Orthogonal Hahn moments. The order ‘m’ Hahn polynomial can be computed from Eq. (11):
The aforementioned Pochhammer symbol ( 
And was simplified further by the Gamma operator in Eq. (13):
The Hahn moments raw values are scaled using a weighting function and a square norm given as in Eq. (14):
Meanwhile, in Eq. (15),
The Hahn moments are computed up to degree-3for the 2-D discrete data as follows in Eq. (16):
The 10 key Hahn moments-based features are represented by ({H}_{00}), ({H}_{01}), ({H}_{10}), ({H}_{11}),({H}_{02}),({H}_{20}),({H}_{12}),({H}_{21}),({H}_{30} mathrm{and} {H}_{03}). Matrix M’ was utilized in computing ten Raw, ten Central and ten Hahn moments for every DNA sample sequence up to degree-3 which later are unified into the miscellany super feature vector (SFV).
The DNA characteristics such as ordered location of the nucleotides in the DNA sequences are of pivotal significance for identification. The relative positioning of nucleotides in any DNA sequence is considered core patterns prevailing the physical features of the DNA sequence. The DNA sequence is represented by D-PRIM in (4 × 4) order. The matrix in Eq. (17) is used to extract position-relative attributes of every nucleotide in the given DNA sequence.
The position occurrence values of nucleotides are represented here using the notation ({N}_{xto y}). Here the indication score of the (y)th position nucleotide is determined using ({N}_{xto y}) with respect to the xth nucleotide first occurrence in the sequence. The nucleotide type ‘(y)’substitutes this score in the biological evolutionary process. The occurrence positional values, in alphabetical order, represented as four native nucleotides. The SD-PRIM matrix is formed with 16 coefficient values obtained after successfully performing computations on position relative incidences. Similarly, SD-PRIM1668and SD-PRIM6468were constructedhaving16 × 16 and 64 × 64 valuable coefficient features respectively. The 2D heatmaps of these matrices are shown in Figs. 1, 2 and 3. These heatmaps are based on the summation of nucleotide, dinucleotide and trinucleotide composition PRIMs.
The heatmap of nucleotide composition based PRIMs.
The heatmap of dinucleotide composition based PRIMs.
The heatmap of trinucleotide composition based PRIMs.
30 raw, central and Hahn moments (10 raw, 10 central & 10 Hahn), up to degree-3, were computed using the 2D SD-PRIM matrix through which 30 features were obtained with 16 unique coefficients and were further incorporated into the miscellany Super Feature Vector (SFV).
It often happens in cellular biology that the same ancestor is responsible for evolving more than one DNA sequence. These cases mostly outcome homologous sequences. The performance of the classifier is hugely affected by these homologous sequences and hence for producing accurate results, sequence similarity searching is reliable and effectively useful. In machine learning, accuracy and efficiency is hugely dependent on the meticulousness and thoroughness of algorithms through which most pertinent features in the data are extracted. The algorithms used in machine learning have the ability to learn and adapt the most obscure patterns embedded in the data while understanding and uncovering them during the learning phase. The procedure followed during the computation of D-PRIM was utilized in computations of D-RPRIM but only with reverse DNA sequence ordering. The position occurrence values of nucleotides are represented here using the notation ({N}_{xto y}). Here the indication score of the (y) thposition nucleotide is determined using ({N}_{xto y}) with respect to the xth nucleotide first occurrence in the sequence. The nucleotide type ‘(y)’ substitutes this score in the biological evolutionary process. The occurrence positional values, in alphabetical order, represented as 4 native nucleotides. This procedure further uncovered hidden patterns for prediction and ambiguities between similar DNA sequences were also alleviated. The 2D matrix D-RPRIM was formed with (4 × 4) order having16unique coefficients. It is defined by Eq. (18):
Similarly, 30 raw, central and Hahn moments (10 raw, 10 central & 10 Hahn), up to degree-3, were computed using the 2D SD-RPRIM matrix through which 30 features were also obtained with 16 unique coefficients and they were also incorporated into the miscellany Super Feature Vector (SFV).
The distribution of occurrence of every nucleotide was used to compute the frequency distribution vector. The frequency distribution vector (FDV) is defined as in Eq. (19):
Here ({rho }_{i}) is the frequency of occurrence of the ith (1 ≤ i ≤ 4) relevant nucleotide. Furthermore, the relative positions of nucleotides in any sequence are highly utilized using these measures. The miscellany Super Feature Vector (SFV) includes these four features from FDV as unique attributes. The violin plots of nucleotide composition and overall frequency normalization is shown in Figs. 4a–d and 5.
(a) The violin plot of nucleotide adenine (A) composition. (b) The violin plot of nucleotide cytosine (C) composition. (c) The violin plot of nucleotide thymine (T) composition. (d) The violin plot of nucleotide guanine (G) composition.
The violin plot of all four nucleotide compositions.
The distributional information of nucleotides was stored using frequency distribution vector which used the hidden patterns features of DNA sequences in relevance to the compositional details. FDV does not have any information regarding relative positional details of relevant nucleotide residues in DNA sequences. This relative positional information was accommodated using D-AAPIV with a length of four critical features associated with four native nucleotides in a DNA sequence. These four critical features from D-AAPIV are also added into the miscellany Super Feature Vector (SFV).
Here ({alpha }_{i}) is any element of D-AAPIV, from DNA sequence ({S}_{j}) having ‘n’ total nucleotides, which can be calculated using Eq. (21):
D-RAAPIV is calculated using the reverse DNA sequence as input with the same method used using D-AAPIV calculations. This vector is calculated to find the deep and hidden features of every sample with respect to reverse relative positional information. D-RAAPIV is formed as the following Eq. (24) using the reversed DNA sequence and generates four valuable features. These four critical features from D-RAAPIV are also added into the miscellany Super Feature Vector (SFV).
Here ({alpha }_{i}) is any element of D-RAAPIV, from DNA sequence ({S}_{j}) having ‘n’ total nucleotides, which can be calculated using Eq. (23):
After calculating all possible features from the aforementioned extraction methods, the Super Feature Vector (SFV)was constructed, for further processing in classification algorithm. The proposed model has used extracted features with more robustness to noise and effective against the sensitive DNA Enhancer sites as shown in Fig. 6. All the combined features efficiently differentiate from Enhancers and Non Enhancer sites.
The feature visualization scatter plot of features extracted and used in the proposed study.
In the past, ensemble learning methods have been applied in many bioinformatics relevant research studies76,77 and have produced highly efficient outcomes in measures of performance. Ensemble learning methods utilize many classifiers in a classification problem with aggregation of their results. The two most commonly used methods are boosting78,79 and bagging80 which perform classifications using trees. In boosting, the trees which are successive, propagate extra weights to points which are predicted incorrectly by the previous classifiers. The weighted vote decides the prediction in the end. Whereas, in bagging, the successive trees do not rely on previous trees, rather, each tree is constructed independently from the data using a bootstrap sample. The simple majority vote decides the prediction in the end.
In bioinformatics and related fields, random forests have grown in popularity as a classification tool. They have also performed admirably in extremely complex data environments. A random sample of the observations, typically a bootstrap sample or a subsample of the original data, is used to build each tree in a random forest. Out-of-bag (OOB) observations are those that are not included in the subsample or the bootstrap sample, respectively. The so-called OOB error can be produced, for instance, by using the OOB observations to estimate the random forest prediction error. The OOB error is frequently used to gauge how well the random forest classifier predicts outcomes and aids in identifying model uncertainties. The OOB error has the benefit of using the entire original sample for both building the random forest classifier and estimating error. In order to add more randomness to bagging, Leo Breiman81 constructed random forests. The random forests changed the construction of the classification trees by adding the construction of each tree from the data using a different bootstrap sample. The splitting of each node, in standard classification trees, is performed by dividing each node equally among all the variables. However, in random forests, the splitting of each node is performed by choosing the best among a subset of predictors which are chosen randomly at that node (Fig. 7 shows the structure of the random forest classifier). As compared to many other classifiers, such as support vector machine, discriminant analysis and neural networks, this counterintuitive strategy perform very well and is robust against overfitting76.
The structure of the random forest classifier.
Scikit-Learn82 library using python was implemented for random forest classifier for fitting the trainings and simulations in our proposed method. The number of trees was increased from the default parameter value of 10 to 100. The number of trees parameter value was optimized to 100 using hyper parameter tuning methods and optimal value for the parameter was searched using the successive halving technique in scikit-learn82 library. The searching space for the parameter “n_estimators” in random forest classifier was (5–500) which was optimized to 100 after successful halving. One of the key findings observed during the experimentation process was that forest with more than 100 trees minimally contribute to the accuracy of the classifier, but can enhance the overall size of the proposed model substantially. Figure 8a illustrates a flowchart to show the overall process of the proposed method.
(a) The Flowchart of the overall proposed method. (b) The OOB error rate stabilization during training estimator trees.
It is frequently asserted that the OOB error is a neutral estimator of the true error rate. Every observation is “out-of-bag” for some of the trees in a random forest because each tree is constructed from a different sample of the original data. Then, only those trees can be used for which the observation was not used in the construction to derive the prediction for the observation. Each observation is given a classification as a result, and the error rate can be calculated using these predictions. The resulting error rate is referred to as OOB error. Breiman81 was the first to propose this process, and it has since gained widespread acceptance as a reliable technique for error estimation in “Random forests”. Each new tree is fitted from a bootstrap sample of the training observations ({z}_{i}= {x}_{i}, {y}_{i}) when training the random forest classifier using bootstrap aggregation. The average error for each ({z}_{i}) calculated using predictions from the trees that do not contain ({z}_{i}) in their respective bootstrap sample is known as the out-of-bag (OOB) error. This makes it possible to fit and validate the random forest classifier while it is being trained. The OOB error is calculated at the addition of each new tree during training, as shown in the plot below. A practitioner can roughly determine the value of n estimators at which the error stabilizes using the resulting Fig. 8b. The scikit-learn82 library was used to process the out of bag error estimation.
This article does not contain any studies involved with human participants or animals performed by any of the authors.
For the assessment and verifications of the model and to analyze its performance, some methods are used to evaluate them. These methods evaluate the classifiers using inspection attributes which are based on the outcomes of classification assessments and estimates.
K-fold cross validation (KFCV) technique is most commonly used by practitioners for estimation of errors in classifications. Also known as rotation estimation, KFCV splits a dataset into ‘K’ folds which are randomly selected and are equal in size approximately. The prediction error of the fitted model is calculated by predicting the kth part of the data which is dependent on other K − 1 parts to fit the model. The error estimates of K from the prediction are combined together using the same procedure for each k = 1, 2, … , K.
Although the generalization performance of any classifier is mostly estimated using unbiased approximations in jackknife tests, two drawbacks exists in this test, firstly, the variance is high as estimates used in all the datasets are very similar to each other, secondly, its calculative expensive as n estimates are required to be computed, and the total number of observations to test is n in the dataset. The fivefold and tenfold cross validation tests are proven to be a good compromise between computational requirements and impartiality.
In the KFCV tests, the selection of ‘K’ is considered as a significant attribute. To testify errors in prediction models, cross validations (K = 5 and K = 10) tests have been used in many research studies. 5-Fold and 10-Fold tests proved to have accurate results in our proposed model and proved to be much better than state-of-the-art methods. These results are listed in Tables 4, 5 and 6.
The problems of binary classifications use metrics such as Accuracy (Acc), Sensitivity (Sn), Specificity (Sp) and Mathew’s Correlation Coefficient (MCC)for measuring the proposed prediction model quality and efficiency. These metrics are defined in the following Eq. (24):
Here true-positives (TP), TN (true-negatives), FP (false-positives) and FN (false-negatives) represent the outcomes from the cross validation tests. Unfortunately, the conventional formulations from the above mentioned metrics in Eq. (24) lack in intuitiveness and due to this fact, understanding these measures especially MCC, many scientists have faced difficulties. To ease this difficulty, the above conventional equations were converted by Xu83 and Feng84 using Chou’s four intuitive equations which used the symbols introduced by Chou85. The symbols that define these equations are; ({mathrm{Y}}^{+}, {mathrm{Y}}^{-},{mathrm{Y}}_{-}^{+}mathrm{ and }{mathrm{Y}}_{+}^{-}).The description of these symbols is defined in Table 2.
From the above correspondence in Table 2, we can define Eq. (25):
From the above correspondence in Table 2, we can define Eq. (26):
The above Eq. (26) has the same meaning as the Eq. (24) but it is more easy to understand and intuitive. Table 3 defines the detail description of these equations.
The set of metrics used in above Table 3 are not applicable to multi-labeled prediction models rather they are only useful for single labeled-systems. A different set of metrics exists for multi-labeled-systems which have been used by various researchers86,87,88. The comparison of existing classifiers with proposed method is mentioned in Tables 4, 5 and 6.
The classification algorithms with their predictions results using benchmark dataset are shown in Tables 4, 5 and 6. iEnhancer-EL89 and iEnhancer-2L26 produced better outcomes using ensemble classifiers and achieved accuracy of 78.03% and 76.89% respectively in which they were successful in predicting strong enhancers with accuracy of 65.03% and 61.93% respectively. Whereas EnhancerPred27 achieved 80.82% accuracy and used SVMs which produced slightly better results in predicting strong enhancers with 62.06% accuracy. Similarly, iEnhancer-2L-Hybrid90 and iEnhancer-5Step29 improved the accuracy results with their prediction model and acquired77.86% and 82.3% accuracies respectively with identifying the strong enhancers with 65.83% and 68.1% accuracies respectively. In contrast, 91.68%and 84.53%accuracy was achieved in predicting enhancers and their strength respectively by the currently proposed method after utilizing obscure features from statistical moments and random forest classifications using 5-Fold cross validation tests (see Table 4 and Fig. 9 for ROCs). Furthermore, tenfold cross-validation test was also conducted using random forest classifier on benchmark dataset and obtained the accuracy results are listed in Table 5. The ROCs of 10-fold cross-validation tests are shown in Figs. 10 and 11. The violin plots of 5 fold cross-validation tests are shown in Fig. 12. In addition to cross validation tests, an independent test was also performed using the independent dataset. The comparison of proposed model and state-of-the-art methods using independent dataset is listed in Table 6 and ROC is shown in Fig. 13. Furthermore, jackknife test was also performed on these datasets. A detailed comparison of some selected machine learning algorithms using jackknife test is mentioned in Table 7. The Precision-Recall (PR) curves for enhancer and their strength prediction is also labeled in Figs. 14 and 15 respectively. The proposed method is based on the feature sets that are evaluated using Hahn moments which are easier for the random forest based classifier to classify the feature vectors in acute time and are very efficient as compared to previous methods which were not able to produce better results on the computational cost of training and testing using classification process.
(a) ROC curve of fivefold cross validation tests for enhancers. (b) ROC curve of fivefold cross validation tests for enhancer strengths.
10 fold test ROCs comparison of classifiers for enhancer site prediction.
(a) ROC curve of tenfold cross validation tests for enhancers using (random forest). (b) ROC curve of tenfold cross validation tests for enhancers strength (random forest).
Violinplot of fivefold cross validation for enhancers (random forest).
The ROCs of state of art methods using independent tests for enhancer prediction.
The PR curves of random forest using jackknife tests for enhancer site prediction.
The PR curves of random forest using jackknife tests for enhancer strength prediction.
As observed in past studies91,92,93,94,95, the development of a web-server is highly significant and useful for building more useful prediction methodologies. Thus, efforts for a user friendly webserver have been made in past72,96,97,98,99 to provide ease to biologists and scientists in drug discovery. The software code which has been developed for the proposed method is accessible at https://github.com/csbioinfopk/enpred which is developed using Python, Scikit-Learn and Flask. The webserver to the current study will be provided for the research community in near future.
In the proposed research, an efficient model for predicting the enhancers and their strength using statistical moments and random forest classifier is developed. In recent past, many methods were proposed to predict enhancers, but our method has proved to be better in accuracy than the existing state-of-the-art methods. Our method achieved accuracies of91.68% and 84.53% for enhancer and strong enhancer classifications using 5 Fold tests on a benchmark dataset which is currently the highest and accurate classification method for prediction of enhancers and their strength.
The Online Supporting Information S1 (https://github.com/csbioinfopk/enpred/blob/master/static/Supp-S1.pdf) provides sequence information of DNA Enhancer and non-Enhancer sites used for training, Online Supporting Information S2 (https://github.com/csbioinfopk/enpred/blob/master/static/Supp-S2.pdf) provides sequence information of DNA Strong enhancer sites and Weak enhancer sites, and Online Supporting Information S3 (https://github.com/csbioinfopk/enpred/blob/master/static/Supp-S3.pdf) provides sequence information of DNA Sample sequences used for Independent Tests. The GitHub repository provide access to all the data necessary with relevant accession numbers to substantiate the study’s findings.
Erwin, G. D. et al. Integrating diverse datasets improves developmental enhancer prediction. PLoS Comput. Biol. 10(6), e1003677–e1003677. https://doi.org/10.1371/journal.pcbi.1003677 (2014).
CAS Article PubMed PubMed Central Google Scholar
Visel, A., Rubin, E. M. & Pennacchio, L. A. Genomic views of distant-acting enhancers. Nature 461(7261), 199–205. https://doi.org/10.1038/nature08451 (2009).
ADS CAS Article PubMed PubMed Central Google Scholar
Sakabe, N. J., Savic, D. & Nobrega, M. A. Transcriptional enhancers in development and disease. Genome Biol. 13(1), 238 (2012).
CAS Article Google Scholar
Heintzman, N. D. & Ren, B. Finding distal regulatory elements in the human genome. Curr. Opin. Genet. Dev. 19(6), 541–549. https://doi.org/10.1016/j.gde.2009.09.006 (2009).
CAS Article PubMed PubMed Central Google Scholar
Blackwood, E. M. & Kadonaga, J. T. Going the distance: A current view of enhancer action. Science 281, 60 (1998).
CAS Article Google Scholar
Pennacchio, L. A., Bickmore, W., Dean, A., Nobrega, M. A. & Bejerano, G. Enhancers: Five essential questions. Nat. Rev. Genet. 14, 288 (2013).
CAS Article Google Scholar
Kulaeva, O. I., Nizovtseva, E. V., Polikanov, Y. S., Ulianov, S. V. & Studitsky, V. M. Distant activation of transcription: Mechanisms of enhancer action. Mol. Cell. Biol. 32(24), 4892–4897. https://doi.org/10.1128/mcb.01127-12 (2012).
CAS Article PubMed PubMed Central Google Scholar
Herz, H.-M. Enhancer deregulation in cancer and other diseases. BioEssays 38(10), 1003–1015. https://doi.org/10.1002/bies.201600106 (2016).
Article PubMed PubMed Central Google Scholar
Zhang, G. et al. DiseaseEnhancer: A resource of human disease-associated enhancer catalog. Nucleic Acids Res. 46(D1), D78–D84. https://doi.org/10.1093/nar/gkx920 (2017).
ADS CAS Article PubMed Central Google Scholar
Corradin, O. & Scacheri, P. C. Enhancer variants: Evaluating functions in common disease. Genome Med. 6, 85 (2014).
Article Google Scholar
Boyd, M. et al. Characterization of the enhancer and promoter landscape of inflammatory bowel disease from human colon biopsies. Nat. Commun. 9, 1661 (2018).
ADS Article Google Scholar
Banerji, J., Rusconi, S. & Schaffner, W. Expression of a β-globin gene is enhanced by remote SV40 DNA sequences. Cell 27(2), 299–308 (1981).
CAS Article Google Scholar
Shlyueva, D., Stampfel, G. & Stark, A. Transcriptional enhancers: From properties to genome-wide predictions. Nat. Rev. Genet. 15(4), 272–286. https://doi.org/10.1038/nrg3682 (2014).
CAS Article PubMed Google Scholar
Heintzman, N. D. et al. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat. Genet. 39(3), 311 (2007).
CAS Article Google Scholar
Jin, F., Li, Y., Ren, B. & Natarajan, R. PU. 1 and C/EBPα synergistically program distinct response to NF-κB activation through establishing monocyte specific enhancers. Proc. Natl. Acad. Sci. 108(13), 5290–5295 (2011).
ADS CAS Article Google Scholar
Kim, T.-K. et al. Widespread transcription at neuronal activity-regulated enhancers. Nature 465(7295), 182 (2010).
ADS CAS Article Google Scholar
Boyle, A. P. et al. High-resolution genome-wide in vivo footprinting of diverse transcription factors in human cells. Genome Res. 21(3), 456–464. https://doi.org/10.1101/gr.112656.110 (2011).
CAS Article PubMed PubMed Central Google Scholar
Visel, A. et al. ChIP-seq accurately predicts tissue-specific activity of enhancers. Nature 457(7231), 854–858. https://doi.org/10.1038/nature07730 (2009).
ADS CAS Article PubMed PubMed Central Google Scholar
Ernst, J. et al. Mapping and analysis of chromatin state dynamics in nine human cell types. Nature 473(7345), 43–49. https://doi.org/10.1038/nature09906 (2011).
ADS CAS Article PubMed PubMed Central Google Scholar
Fernández, M. & Miranda-Saavedra, D. Genome-wide enhancer prediction from epigenetic signatures using genetic algorithm-optimized support vector machines. Nucleic Acids Res. 40(10), e77–e77. https://doi.org/10.1093/nar/gks149 (2012).
CAS Article PubMed PubMed Central Google Scholar
Firpi, H. A., Ucar, D. & Tan, K. Discover regulatory DNA elements using chromatin signatures and artificial neural network. Bioinformatics 26(13), 1579–1586. https://doi.org/10.1093/bioinformatics/btq248 (2010).
CAS Article PubMed PubMed Central Google Scholar
Kleftogiannis, D., Kalnis, P. & Bajic, V. B. DEEP: A general computational framework for predicting enhancers. Nucleic Acids Res. 43(1), e6. https://doi.org/10.1093/nar/gku1058 (2015).
CAS Article PubMed Google Scholar
Rajagopal, N. et al. RFECS: A random-forest based algorithm for enhancer identification from chromatin state. PLoS Comput. Biol. 9(3), e1002968–e1002968. https://doi.org/10.1371/journal.pcbi.1002968 (2013).
CAS Article PubMed PubMed Central Google Scholar
Bu, H., Gan, Y., Wang, Y., Zhou, S. & Guan, J. A new method for enhancer prediction based on deep belief network. BMC Bioinform. 18, 418 (2017).
Article Google Scholar
Yang, B. et al. BiRen: Predicting enhancers with a deep-learning-based model using the DNA sequence alone. Bioinformatics 33(13), 1930–1936 (2017).
CAS Article Google Scholar
Liu, B., Fang, L., Long, R., Lan, X. & Chou, K. C. iEnhancer-2L: A two-layer predictor for identifying enhancers and their strength by pseudo k-tuple nucleotide composition. Bioinformatics 32(3), 362–369. https://doi.org/10.1093/bioinformatics/btv604 (2016).
CAS Article PubMed Google Scholar
Jia, C. & He, W. EnhancerPred: A predictor for discovering enhancers based on the combination and selection of multiple features. Sci. Rep. https://doi.org/10.1038/srep38741 (2016).
Article PubMed PubMed Central Google Scholar
He, W. & Jia, C. EnhancerPred2.0: Predicting enhancers and their strength based on position-specific trinucleotide propensity and electron-ion interaction potential feature selection. Mol. BioSyst. 13(4), 767–774. https://doi.org/10.1039/c7mb00054e (2017).
CAS Article PubMed Google Scholar
Le, N. Q. K. et al. iEnhancer-5Step: Identifying enhancers using hidden information of DNA sequences via Chou’s 5-step rule and word embedding. Anal. Biochem. 571, 53–61. https://doi.org/10.1016/j.ab.2019.02.017 (2019).
CAS Article PubMed Google Scholar
Yang, H., Wang, S. & Xia, X. iEnhancer-RD: Identification of enhancers and their strength using RKPK features and deep neural networks. Anal. Biochem. 630, 114318. https://doi.org/10.1016/j.ab.2021.114318 (2021).
CAS Article PubMed Google Scholar
Zhang, T.-H., Flores, M. & Huang, Y. ES-ARCNN: Predicting enhancer strength by using data augmentation and residual convolutional neural network. Anal. Biochem. 618, 114120. https://doi.org/10.1016/j.ab.2021.114120 (2021).
CAS Article PubMed Google Scholar
Yang, R., Wu, F., Zhang, C. & Zhang, L. iEnhancer-GAN: A deep learning framework in combination with word embedding and sequence generative adversarial net to identify enhancers and their strength. Int. J. Mol. Sci. 22(7), 3589. https://doi.org/10.3390/ijms22073589 (2021).
CAS Article PubMed PubMed Central Google Scholar
Cai, L. et al. iEnhancer-XG: Interpretable sequence-based enhancers and their strength predictor. Bioinformatics 37(8), 1060–1067. https://doi.org/10.1093/bioinformatics/btaa914 (2021).
CAS Article PubMed Google Scholar
Lyu, Y. et al. iEnhancer-KL: A novel two-layer predictor for identifying enhancers by position specific of nucleotide composition. IEEE/ACM Trans. Comput. Biol. Bioinf. 18(6), 2809–2815. https://doi.org/10.1109/TCBB.2021.3053608 (2021).
CAS Article Google Scholar
Le, N. Q. K., Ho, Q.-T., Nguyen, T.-T.-D. & Ou, Y.-Y. ‘A transformer architecture based on BERT and 2D convolutional neural network to identify DNA enhancers from sequence information. Brief Bioinform. 22(5), bbab005. https://doi.org/10.1093/bib/bbab005 (2021).
CAS Article PubMed Google Scholar
Liang, Y., Zhang, S., Qiao, H. & Cheng, Y. iEnhancer-MFGBDT: Identifying enhancers and their strength by fusing multiple features and gradient boosting decision tree. Math. Biosci. Eng. 18(6), 8797–8814. https://doi.org/10.3934/mbe.2021434 (2021).
MathSciNet Article PubMed Google Scholar
Nguyen, Q. H. et al. iEnhancer-ECNN: Identifying enhancers and their strength using ensembles of convolutional neural networks. BMC Genomics 20(Suppl 9), 951. https://doi.org/10.1186/s12864-019-6336-3 (2019).
CAS Article PubMed PubMed Central Google Scholar
Tan, K. K., Le, N. Q. K., Yeh, H. Y. & Chua, M. C. H. ‘Ensemble of deep recurrent neural networks for identifying enhancers via dinucleotide physicochemical properties. Cells 8(7), 767. https://doi.org/10.3390/cells8070767 (2019).
CAS Article PubMed Central Google Scholar
Fu, L., Niu, B., Zhu, Z., Wu, S. & Li, W. CD-HIT: Accelerated for clustering the next-generation sequencing data. Bioinformatics 28(23), 3150–3152. https://doi.org/10.1093/bioinformatics/bts565 (2012).
CAS Article PubMed PubMed Central Google Scholar
Chou, K.-C. Impacts of bioinformatics to medicinal chemistry. Med. Chem. 11(3), 218–234. https://doi.org/10.2174/1573406411666141229162834 (2015).
CAS Article PubMed Google Scholar
Chou, K. C. Prediction of protein cellular attributes using pseudo-amino acid composition. Proteins Struct. Funct. Genet. 43(3), 246–255. https://doi.org/10.1002/prot.1035 (2001).
CAS Article PubMed Google Scholar
Cao, D.-S., Xu, Q.-S. & Liang, Y.-Z. propy: A tool to generate various modes of Chou’s PseAAC. Bioinformatics 29(7), 960–962. https://doi.org/10.1093/bioinformatics/btt072 (2013).
CAS Article PubMed Google Scholar
Du, P., Wang, X., Xu, C. & Gao, Y. PseAAC-Builder: A cross-platform stand-alone program for generating various special Chou’s pseudo-amino acid compositions. Anal. Biochem. 425, 117–119. https://doi.org/10.1016/j.ab.2012.03.015 (2012).
CAS Article PubMed Google Scholar
Du, P., Gu, S. & Jiao, Y. PseAAC-general: Fast building various modes of general form of Chou’s pseudo-amino acid composition for large-scale protein datasets. Int. J. Mol. Sci. 15, 3495 (2014).
CAS Article Google Scholar
Chou, K. C. Some remarks on protein attribute prediction and pseudo amino acid composition. J. Theor. Biol. 273(1), 236–247. https://doi.org/10.1016/j.jtbi.2010.12.024 (2011).
ADS MathSciNet CAS Article PubMed MATH Google Scholar
Chou, K.-C. Pseudo amino acid composition and its applications in bioinformatics, proteomics and system biology. Curr. Proteom. 6(4), 262–274. https://doi.org/10.2174/157016409789973707 (2009).
CAS Article Google Scholar
Chen, W., Lei, T. Y., Jin, D. C., Lin, H. & Chou, K. C. PseKNC: A flexible web server for generating pseudo K-tuple nucleotide composition. Anal. Biochem. 456(1), 53–60. https://doi.org/10.1016/j.ab.2014.04.001 (2014).
CAS Article PubMed Google Scholar
Chen, W., Lin, H. & Chou, K. C. Pseudo nucleotide composition or PseKNC: An effective formulation for analyzing genomic sequences. Mol. BioSyst. 11(10), 2620–2634. https://doi.org/10.1039/c5mb00155b (2015).
CAS Article PubMed Google Scholar
Liu, B., Yang, F., Huang, D.-S. & Chou, K.-C. iPromoter-2L: A two-layer predictor for identifying promoters and their types by multi-window-based PseKNC. Bioinformatics 34(1), 33–40. https://doi.org/10.1093/bioinformatics/btx579 (2017).
CAS Article Google Scholar
Liu, B. et al. Pse-in-One: A web server for generating various modes of pseudo components of DNA, RNA, and protein sequences. Nucleic Acids Res. 43(W1), W65–W71. https://doi.org/10.1093/nar/gkv458 (2015).
CAS Article PubMed PubMed Central Google Scholar
Liu, B., Wu, H. & Chou, K.-C. Pse-in-One 2.0: An improved package of web servers for generating various modes of pseudo components of DNA, RNA, and protein sequences. Nat. Sci. 09(04), 67–91. https://doi.org/10.4236/ns.2017.94007 (2017).
CAS Article Google Scholar
Liu, B., Long, R. & Chou, K. C. IDHS-EL: Identifying DNase i hypersensitive sites by fusing three different modes of pseudo nucleotide composition into an ensemble learning framework. Bioinformatics 32(16), 2411–2418. https://doi.org/10.1093/bioinformatics/btw186 (2016).
Article PubMed Google Scholar
Papademetriou, R. C. ‘Reconstructing with moments. Proc. Int. Conf. Pattern Recogn. 3, 476–480. https://doi.org/10.1109/ICPR.1992.202028 (1992).
Article Google Scholar
Butt, A. H., Khan, S. A., Jamil, H., Rasool, N. & Khan, Y. D. A prediction model for membrane proteins using moments based features. Biomed. Res. Int. 2016, 1–7. https://doi.org/10.1155/2016/8370132 (2016).
CAS Article Google Scholar
Butt, A. H., Rasool, N. & Khan, Y. D. A treatise to computational approaches towards prediction of membrane protein and its subtypes. J. Membr. Biol. 250(1), 55–76. https://doi.org/10.1007/s00232-016-9937-7 (2017).
CAS Article PubMed Google Scholar
Butt, A. H., Rasool, N. & Khan, Y. D. Predicting membrane proteins and their types by extracting various sequence features into Chou’s general PseAAC. Mol. Biol. Rep. 45(6), 2295–2306. https://doi.org/10.1007/s11033-018-4391-5 (2018).
CAS Article PubMed Google Scholar
Butt, A. H., Rasool, N. & Khan, Y. D. Prediction of antioxidant proteins by incorporating statistical moments based features into Chou’s PseAAC. J. Theor. Biol. 473, 1–8. https://doi.org/10.1016/j.jtbi.2019.04.019 (2019).
ADS CAS Article PubMed MATH Google Scholar
Butt, A. H. & Khan, Y. D. CanLect-Pred: A cancer therapeutics tool for prediction of target cancer lectins using experiential annotated proteomic sequences. IEEE Access https://doi.org/10.1109/ACCESS.2019.2962002 (2020).
Article Google Scholar
Khan, Y. D., Khan, N. S., Naseer, S. & Butt, A. H. iSUMOK-PseAAC: Prediction of lysine sumoylation sites using statistical moments and Chou’s PseAAC. PeerJ 9, e11581. https://doi.org/10.7717/peerj.11581 (2021).
CAS Article PubMed PubMed Central Google Scholar
Khan, S. A., Khan, Y. D., Ahmad, S. & Allehaibi, K. H. N-MyristoylG-PseAAC: Sequence-based prediction of N-Myristoyl glycine sites in proteins by integration of PseAAC and statistical moments. Lett. Org. Chem. 16(3), 226–234. https://doi.org/10.2174/1570178616666181217153958 (2019).
CAS Article Google Scholar
Amanat, S., Ashraf, A., Hussain, W., Rasool, N. & Khan, Y. D. Identification of lysine carboxylation sites in proteins by integrating statistical moments and position relative features via general PseAAC. Curr. Bioinform. 15(5), 396–407. https://doi.org/10.2174/1574893614666190723114923 (2020).
CAS Article Google Scholar
Mahmood, M. K., Ehsan, A., Khan, Y. D. & Chou, K.-C. iHyd-LysSite (EPSV): Identifying hydroxylysine sites in protein using statistical formulation by extracting enhanced position and sequence variant feature technique. Curr. Genomics 21(7), 536–545. https://doi.org/10.2174/1389202921999200831142629 (2020).
CAS Article PubMed PubMed Central Google Scholar
Khan, Y. D., Khan, S. A., Ahmad, F. & Islam, S. Iris recognition using image moments and k-Means algorithm. Sci. World J. 2014, 1–9. https://doi.org/10.1155/2014/723595 (2014).
Article Google Scholar
Zhou, J., Shu, H., Zhu, H., Toumoulin, C., & Luo, L. Image analysis by discrete orthogonal Hahn moments. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 3656. 524–531. https://doi.org/10.1007/11559573_65 (LNCS, 2005).
Zhu, H., Shu, H., Zhou, J., Luo, L. & Coatrieux, J. L. Image analysis by discrete orthogonal dual Hahn moments. Pattern Recogn. Lett. 28(13), 1688–1704. https://doi.org/10.1016/j.patrec.2007.04.013 (2007).
ADS Article Google Scholar
Yap, P. T., Paramesran, R. & Ong, S. H. Image analysis using Hahn moments. IEEE Trans. Pattern Anal. Mach. Intell. 29(11), 2057–2062. https://doi.org/10.1109/TPAMI.2007.70709 (2007).
Article PubMed Google Scholar
Goh, H.-A., Chong, C.-W., Besar, R., Abas, F. S. & Sim, K.-S. Translation and scale invariants of Hahn moments. Int. J. Image Graph. 09(02), 271–285. https://doi.org/10.1142/s0219467809003435 (2009).
Article Google Scholar
Alghamdi, W., Alzahrani, E., Ullah, M. Z. & Khan, Y. D. 4mC-RF: Improving the prediction of 4mC sites using composition and position relative features and statistical moment. Anal. Biochem. 633, 114385. https://doi.org/10.1016/j.ab.2021.114385 (2021).
CAS Article PubMed Google Scholar
Malebary, S. J., ur Rehman, M. S. & Khan, Y. D. iCrotoK-PseAAC: Identify lysine crotonylation sites by blending position relative statistical features according to the Chou’s 5-step rule. PLoS ONE 14(11), 0223993. https://doi.org/10.1371/journal.pone.0223993 (2019).
CAS Article Google Scholar
Shah, A. A. & Khan, Y. D. Identification of 4-carboxyglutamate residue sites based on position based statistical feature and multiple classification. Sci. Rep. 10(1), 16913. https://doi.org/10.1038/s41598-020-73107-y (2020).
ADS CAS Article PubMed PubMed Central Google Scholar
Ilyas, S. et al. iMethylK_pseAAC: Improving accuracy of lysine methylation sites identification by incorporating statistical moments and position relative features into general PseAAC via Chou’s 5-steps rule. Curr. Genom. 20(4), 275–292. https://doi.org/10.2174/1389202920666190809095206 (2019).
CAS Article Google Scholar
Awais, M. et al. iPhosH-PseAAC: Identify phosphohistidine sites in proteins by blending statistical moments and position relative features according to the Chou’s 5-step rule and general pseudo amino acid composition. IEEE/ACM Trans. Comput. Biol. Bioinform. https://doi.org/10.1109/TCBB.2019.2919025 (2019).
Article Google Scholar
Barukab, O., Khan, Y. D., Khan, S. A. & Chou, K.-C. iSulfoTyr-PseAAC: Identify tyrosine sulfation sites by incorporating statistical moments via Chou’s 5-steps rule and pseudo components. Curr. Genomics 20(4), 306–320. https://doi.org/10.2174/1389202920666190819091609 (2019).
CAS Article PubMed PubMed Central Google Scholar
Akmal, M. A., Rasool, N. & Khan, Y. D. Prediction of N-linked glycosylation sites using position relative features and statistical moments. PLoS ONE 12(8), e0181966–e0181966. https://doi.org/10.1371/journal.pone.0181966 (2017).
CAS Article PubMed PubMed Central Google Scholar
Khan, Y. D., Batool, A., Rasool, N., Khan, S. A. & Chou, K.-C. Prediction of nitrosocysteine sites using position and composition variant features. Lett. Org. Chem. 16(4), 283–293. https://doi.org/10.2174/1570178615666180802122953 (2018).
CAS Article Google Scholar
Tyryshkina, A., Coraor, N. & Nekrutenko, A. Predicting runtimes of bioinformatics tools based on historical data: Five years of Galaxy usage. Bioinformatics 35(18), 3453–3460. https://doi.org/10.1093/bioinformatics/btz054 (2019).
CAS Article PubMed PubMed Central Google Scholar
Simidjievski, N., Todorovski, L. & Džeroski, S. Modeling dynamic systems with efficient ensembles of process-based models. PLoS ONE 11, 4. https://doi.org/10.1371/journal.pone.0153507 (2016).
CAS Article Google Scholar
Freund, Y. & Schapire, R. E. A decision-theoretic generalization of on-line learning and an application to boosting. Lect. Notes Comput. Sci. (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 904(1), 23–37. https://doi.org/10.1006/jcss.1997.1504 (1995).
Article Google Scholar
Schapire, R. E. Theoretical, views of boosting and applications. Lect. Notes Comput. Sci. (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 1720, 13–25. https://doi.org/10.1007/3-540-46769-6_2 (1999).
MathSciNet Article Google Scholar
Breiman, L. Bagging predictors. Mach. Learn. 24(2), 123–140. https://doi.org/10.1007/bf00058655 (1996).
Article MATH Google Scholar
Breiman, L. Random forests. Mach. Learn. 45(1), 5–32. https://doi.org/10.1023/A:1010933404324 (2001).
Article MATH Google Scholar
Pedregosa, F. et al. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011).
MathSciNet MATH Google Scholar
Xu, Y., Shao, X. J., Wu, L. Y., Deng, N. Y. & Chou, K. C. ISNO-AAPair: Incorporating amino acid pairwise coupling into PseAAC for predicting cysteine S-nitrosylation sites in proteins. PeerJ 2013(1), e171–e171. https://doi.org/10.7717/peerj.171 (2013).
CAS Article Google Scholar
Feng, P. M., Ding, H., Chen, W. & Lin, H. Naïve bayes classifier with feature selection to identify phage virion proteins. Comput. Math. Methods Med. 2013, 1–6. https://doi.org/10.1155/2013/530696 (2013).
Article MATH Google Scholar
Chou, K. C. Prediction of signal peptides using scaled window. Peptides 22(12), 1973–1979. https://doi.org/10.1016/S0196-9781(01)00540-X (2001).
CAS Article PubMed Google Scholar
Xiao, X., Wang, P., Lin, W. Z., Jia, J. H. & Chou, K. C. IAMP-2L: A two-level multi-label classifier for identifying antimicrobial peptides and their functional types. Anal. Biochem. 436(2), 168–177. https://doi.org/10.1016/j.ab.2013.01.019 (2013).
CAS Article PubMed Google Scholar
Xiao, X., Wu, Z. C. & Chou, K. C. iLoc-Virus: A multi-label learning classifier for identifying the subcellular localization of virus proteins with both single and multiple sites. J. Theor. Biol. 284(1), 42–51. https://doi.org/10.1016/j.jtbi.2011.06.005 (2011).
ADS CAS Article PubMed MATH Google Scholar
Lin, W. Z., Fang, J. A., Xiao, X. & Chou, K. C. ILoc-Animal: A multi-label learning classifier for predicting subcellular localization of animal proteins. Mol. BioSyst. 9(4), 634–644. https://doi.org/10.1039/c3mb25466f (2013).
CAS Article PubMed Google Scholar
Liu, B., Li, K., Huang, D. S. & Chou, K. C. IEnhancer-EL: Identifying enhancers and their strength with ensemble learning approach. Bioinformatics 34(22), 3835–3842. https://doi.org/10.1093/bioinformatics/bty458 (2018).
CAS Article PubMed Google Scholar
Tahir, M., Hayat, M. & Khan, S. A. A two-layer computational model for discrimination of enhancer and their types using hybrid features pace of pseudo K-tuple nucleotide composition. Arab. J. Sci. Eng. 43(12), 6719–6727. https://doi.org/10.1007/s13369-017-2818-2 (2018).
CAS Article Google Scholar
Cheng, X., Xiao, X. & Chou, K. C. pLoc_bal-mGneg: Predict subcellular localization of Gram-negative bacterial proteins by quasi-balancing training dataset and general PseAAC. J. Theor. Biol. 458, 92–102. https://doi.org/10.1016/j.jtbi.2018.09.005 (2018).
ADS CAS Article PubMed MATH Google Scholar
Chou, K.-C. Proposing pseudo amino acid components is an important milestone for proteome and genome analyses. Int. J. Pept. Res. Ther. https://doi.org/10.1007/s10989-019-09910-7 (2019).
Article Google Scholar
Liu, B., Wu, H., Zhang, D., Wang, X. & Chou, K. C. Pse-Analysis: A python package for DNA/RNA and protein/ peptide sequence analysis based on pseudo components and kernel methods. Oncotarget 8(8), 13338–13343. https://doi.org/10.18632/oncotarget.14524 (2017).
Article PubMed PubMed Central Google Scholar
Liu, Z. et al. pRNAm-PC: Predicting N6-methyladenosine sites in RNA sequences via physical-chemical properties. Anal. Biochem. 497, 60–67. https://doi.org/10.1016/j.ab.2015.12.017 (2016).
CAS Article PubMed Google Scholar
Feng, P. et al. iDNA6mA-PseKNC: Identifying DNA N 6-methyladenosine sites by incorporating nucleotide physicochemical properties into PseKNC. Genomics 111(1), 96–102. https://doi.org/10.1016/j.ygeno.2018.01.005 (2019).
CAS Article PubMed Google Scholar
Hussain, W., Khan, Y. D., Rasool, N., Khan, S. A. & Chou, K. C. SPrenylC-PseAAC: A sequence-based model developed via Chou’s 5-steps rule and general PseAAC for identifying S-prenylation sites in proteins. J. Theor. Biol. 468, 1–11. https://doi.org/10.1016/j.jtbi.2019.02.007 (2019).
ADS CAS Article PubMed MATH Google Scholar
Ghauri, A. W., Khan, Y. D., Rasool, N., Khan, S. A. & Chou, K.-C. pNitro-Tyr-PseAAC: Predict nitrotyrosine sites in proteins by incorporating five features into Chou’s general PseAAC. Curr. Pharm. Des. 24(34), 4034–4043. https://doi.org/10.2174/1381612825666181127101039 (2018).
CAS Article PubMed Google Scholar
Khan, Y. D. et al. pSSbond-PseAAC: Prediction of disulfide bonding sites by integration of PseAAC and statistical moments. J. Theor. Biol. 463, 47–55. https://doi.org/10.1016/j.jtbi.2018.12.015 (2019).
ADS CAS Article PubMed MATH Google Scholar
Khan, Y. D., Rasool, N., Hussain, W., Khan, S. A. & Chou, K. C. iPhosY-PseAAC: Identify phosphotyrosine sites by incorporating sequence statistical moments into PseAAC. Mol. Biol. Rep. 45(6), 2501–2509. https://doi.org/10.1007/s11033-018-4417-z (2018).
CAS Article PubMed Google Scholar
Download references
The researchers would like to thank the Deanship of Scientific Research, Qassim University for funding the publication of this project.
Department of Computer Science, School of Systems and Technology, University of Management and Technology, Lahore, Pakistan
Ahmad Hassan Butt & Yaser Daanial Khan
Department of Computer, College of Science and Arts in Ar Rass, Qassim University, Ar Rass, Saudi Arabia
Tamim Alkhalifah & Fahad Alturise
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
All authors with equal collaboration, conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the paper, and approved the final draft.
Correspondence to Tamim Alkhalifah.
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and Permissions
Butt, A.H., Alkhalifah, T., Alturise, F. et al. A machine learning technique for identifying DNA enhancer regions utilizing CIS-regulatory element patterns. Sci Rep 12, 15183 (2022). https://doi.org/10.1038/s41598-022-19099-3
Download citation
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-022-19099-3
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
Advertisement
Advanced search
© 2022 Springer Nature Limited
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
L-Acoustics L-ISA Studio 2.4 – A Real-World Review, Part 1 – Mix
Update your browser to view this website correctly. Update my browser now
×
Home Technology
Veteran live sound pro Steve La Cerra dives into L-Acoustics’ L-ISA Studio 2.4 software to create an immersive sound experience.
By Steve La Cerra
It seems like immersive audio is exploding around us. In addition to commercial applications such as theaters, theme parks, gaming and interactive gallery displays, immersive is also gaining popularity in music applications, with artists like Katy Perry, Bon Iver, Ariana Grande, Coldplay, Mark Knopfler, Billie Eilish, Post Malone and the Berlin Philharmonic, among many others, exploring the medium.
The challenge in creating new content is: How can you create an immersive sound experience if you don’t have access to an immersive playback system?
Enter L-Acoustics, one of the first live sound pro-audio manufacturers to dive into the world of immersive sound by introducing L-ISA Immersive Hyperreal Sound, a combination of innovative loudspeaker configurations and object-based mixing and processing tools. L-ISA increases the deployment of loudspeakers from the current 2.1 stereo to a minimum of 5.1 and up to a 360-degree field, enabling sounds to be localized to any area of the sound field.
Until recently, the L-ISA Processor II hardware processor— capable of rendering up to 128 outputs in a 2D or 3D layout—was required to create immersive playback. Earlier this year, L-Acoustics announced that its L-ISA Studio software is available free of charge (from the L-Acoustics website) and is compatible with just about any Mac computer. The application provides an audio engine that includes binaural processing, allowing a user to create and monitor spatialized audio using headphones, with optional head tracking (I used a Head Tracker 1 from Supperware, recommended by L-Acoustics).
L-ISA Studio accepts up to 96 audio sources and can send 12 outputs to any sound card. Content creators can do “preproduction” immersive mixing using headphones, then port the mix to an immersive P.A. system in one of two ways: Use L-ISA Studio software with a hardware audio interface to output 12 discrete audio channels; or, scale the file up to a larger system using the L-ISA Processor II. In essence, it can be used for any kind of playback—and not just live show production.
The L-ISA Studio download contains three components: L-ISA Controller, L-ISA Audio Bridge and L-ISA Processor. L-ISA Controller serves as the user interface where you create, view and modify a soundcscape. L-ISA Processor is the audio engine that processes audio inputs and outputs, and L-ISA Bridge is a software link between the L-ISA Processor and your DAW. When installed properly, L-ISA Audio Bridge appears as a virtual hardware I/O in your DAW and will route multichannel audio internally from your DAW into L-ISA Processor without need for an external audio processor.
When you start L-ISA Processor for the first time, you must allow access to the computer microphone in order for audio to pass. Configuring the system is a bit tricky, and while printed documentation is sparse, there are online tutorials that are helpful.
When L-ISA Processor is opened, the input defaults to L-ISA Audio Bridge with 96 channels; these channels will appear as inputs and outputs in your DAW. The output device can be the computer’s internal speakers (which will be defeated when you plug in your cans) or an external audio interface, which can either feed headphones or a surround mixing system of up to 12 channels. It’s worth noting that the binaural engine still works if you use the headphone jack on an audio interface.
It would take pages to detail the configuration process, but there are a few things worth mentioning:
Once L-ISA Processor and L-ISA Controller have been configured, you can import or create a speaker layout, which defines the physical or binaurally simulated speaker layout. Thankfully, L-Acoustics has provided 5.1, 7.1, 7.1.2 Atmos, 12.0 Binaural and 18.1.4 templates, so you don’t have to manually add each speaker, though you can do so by typing values for the X, Y and Z axes. A speaker layout can contain up to 57 (!) speakers, and even though not all of those can be output when using L-ISA Studio, they still define the layout and contribute to the binaural simulation.
Speaker layouts may also be imported from L-Acoustics’ Soundvision (system modeling software) files. The Soundscape menu is where you’ll see an overhead graphic representation of the speaker layout. Speaker data in an existing session can be changed without destroying the Sources you have already created, making it easy to create sessions for different playback formats.
When a template is opened (or when creating a speaker layout from scratch), outputs will not yet have been assigned to the speakers. This is done in either the Speakers or Routing tabs under the Settings menu. It’s probably best to visit the Routing tab first because you’ll find that the first two channels (BIN-L and BIN-R, binaural), have not been assigned outputs— which means you can’t hear them. Assign these to outputs 1 and 2, respectively, and you should be able to hear audio through your headphones.
A Source is a path from the DAW through the L-ISA Audio Bridge to L-ISA Controller. Each track from the DAW must be assigned to a Source (i.e., an output) in order to be audible, and you can create up to 96 Sources, which you’ll see in the Soundscape. A Source can be mono or stereo, and Sources may be paired, grouped and color-coded for easy management.
Maximum spatial control is achieved by creating a unique Source for every element in a mix (great for controlling sound effects). There are five parameters per Source, displayed at the bottom of the Soundscape window: Pan, Width, Distance, Elevation and Aux send (which can be used to create a downmix or a subwoofer feed). Stereo sources also have a Pan Spread parameter. These can be adjusted using sliders, by typing a value, or by dragging an object in the Soundscape. Sources may be soloed, and there’s a “Live On Air” feature to prevent accidentally soloing a Source during a show.
An easy way to become familiar with the capabilities of L-ISA Studio is by creating and naming a Source for each instrument, using stereo Sources for things like drums, keys and backing vocals. I was able to build mixes by placing Sources in the Soundscape wherever I wanted them to be heard. With the head tracker enabled and the binaural engine turned on, moving your head is freaky: You’ll hear the panorama move based upon the position of your head; if you face the rear of the Soundscape, the binaural processing provides a pretty good illusion that you’ve turned your back to the stage.
The L-ISA Studio audio engine has the capability of generating several different high-quality, multichannel reverb algorithms. You don’t send signal to these reverbs in the traditional sense. L-ISA Studio generates reverb based upon the position of a Source in the Soundscape. As distance from the listening position is increased, so is the reverb—and the reverb algorithms are of very high quality. For each Source, you can turn the following reverb parameters on or off: Direct, Early, Cluster or Late Reflections, Distance Gain, or HF Attenuation—which facilitates creating Sources that have no direct sound and can be moved discretely to specific areas of the Soundscape.
All of this would be very cool (though a bit boring) if it was static—but it isn’t. L-ISA Studio provides snapshot automation that can be fired using OSC, MTC or DeskLink (I used MTC). New Snapshots are created, saved and edited in the Snapshots List under the Soundscape menu, but oddly, you can delete a Snapshot only from the Snapshots menu. Parameters include name, OSC ID, MIDI Program Change number, time code capture and crossfade time, making it easy to create complex changes in the Soundscape.
Dynamic movement of Sources can be automated using the included L-ISA Group Control and L-ISA Source Control plug-ins (VST3, AAX, AU and AAXDSP). A Source or Group can be moved by grabbing it and dragging it in the plugin, as well as by moving the sliders for Pan, Width, Distance Elevation and Aux Send (make sure that plug-in and snapshot control is enabled in the Control tab). Automation moves can be layered, so I was able to move a Source across the panorama on a first pass using the Soundscape graphic, and then add changes in elevation by grabbing the Elevation slider on a second automation pass, which produces really eerie results.
You may only enable either plug-in automation or snapshot automation for a Source parameter, but you can “mix and match” automation—simultaneously controlling some Source parameters using the plug-in and others using Snapshot automation. This results in a tremendous amount of control, providing you plan carefully which type of automation is to be used for each Source parameter.
Steve La Cerra
Audio Bridge ⋅ Concert Sound ⋅ Controller ⋅ immersive ⋅ L Acoustics ⋅ L-ISA ⋅ L-ISA Studio ⋅ Live Sound ⋅ real-world review ⋅ Review ⋅ Steve La Cerra ⋅ venue sound
© 2022 MIX is part of Future plc, an international media group and leading digital publisher. Visit our corporate site.
Worship Software Market – Forecast (2022 – 2030) – Fighting Hawks Magazine
Worship Software Market – Forecast (2022 – 2030) Fighting Hawks Magazine
source
The top iPhone and iPad apps on App Store – Bay News 9
Get the best experience and stay connected to your community with our Spectrum News app. Learn More
Continue in Browser
Get hyperlocal forecasts, radar and weather alerts.
Please enter a valid zipcode.
Save
App Store
Top Paid iPhone Apps:
1. Minecraft, Mojang
2. HotSchedules, HotSchedules
3. Geometry Dash, RobTop Games AB
4. Bloons TD 6, Ninja Kiwi
5. Heads Up!, Warner Bros.
6. Shadowrocket, Shadow Launch Technology Limited
7. Rovio Classics: AB, Rovio Entertainment Oyj
8. Procreate Pocket, Savage Interactive Pty Ltd
9. Monopoly – Classic Board Game, Marmalade Game Studio
10. Papa’s Freezeria To Go!, Flipline Studios
Top Free iPhone Apps:
1. BeReal. Your friends for real., BeReal
2. X-HERO, BINGCHUAN NETWORK (HONG KONG) COMPANY LIMITED
3. YouTube: Watch, Listen, Stream,
4. Google LLC
5. TikTok, TikTok Ltd.
6. Instagram, Instagram, Inc.
7. Google, Google LLC
8. Snapchat, Snap, Inc.
9. Facebook, Meta Platforms, Inc.
10. WhatsApp Messenger, WhatsApp Inc.
11. Gmail – Email by Google, Google LLC
Top Paid iPad Apps:
1. Procreate, Savage Interactive Pty Ltd
2. Minecraft, Mojang
3. Poppy Playtime Chapter 2, MOBGames
4. Bloons TD 6, Ninja Kiwi
5. Geometry Dash, RobTop Games AB
6. Shadowrocket, Shadow Launch Technology Limited
7. Poppy Playtime Chapter 1, MOBGames
8. forScore, forScore, LLC
9. Toca Kitchen 2, Toca Boca AB
10. Stardew Valley, ConcernedApe
Top Free iPad Apps:
1. YouTube: Watch, Listen, Stream, Google LLC
2. Google Chrome, Google LLC
3. Survivor!.io, HABBY
4. Netflix, Netflix, Inc.
5. Stumble Guys, Kitka Games
6. Roblox, Roblox Corporation
7. Disney+, Disney
8. GoodNotes 5, Time Base Technology Limited
9. TikTok, TikTok Ltd.
10. Amazon Prime Video, AMZN Mobile LLC
Copyright 2022 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed without permission.
Big Thief Set Extensive World Tour for 2023 – Rolling Stone
By Larisha Paul
Big Thief’s already expansive tour in support of its latest studio album, Dragon New Warm Mountain I Believe in You, has been expanded with the addition of dozens of dates throughout North America and Europe.
Following the completion of already scheduled shows in South Korea, Japan, Australia, and New Zealand, the indie-folk band will return stateside to begin the North American leg. The portion of shows opens with a performance at Higher Ground in Burlington, Vermont, on Jan. 31 and makes stops in Philadelphia, Tulsa, Nashville, Houston, Miami, Atlanta, New Orleans, and more. It wraps with a grand finale at New York’s Radio City Music Hall on March 2.
The Dragon tour picks back up on Apr. 5 in Gateshead, UK. During the European leg, Big Thief will perform at venues in Manchester, London, Rotterdam, Antwerp, Barcelona, Madrid, and more before concluding in Lisbon on Apr. 29.
General sale for the newly added tour dates begins on Sept. 16 at 10 a.m. local time.
The hefty trek on the road finds Big Thief performing at some of the biggest venues of their career, an achievement that matches the volume of music they’ve created and shared over the last few years while cultivating a stronger connection with their audience.
“I’ve been getting notes that the music has been helping people love and accept themselves in dark times, and to me, that right there is what I want to do,” Big Thief’s Adrianne Lenker told Rolling Stone earlier this year. “I don’t want to get to a certain level in my career or be seen as this or that. I just want people to come back to themselves and be able to accept, love or forgive themselves more fully. I want my music to be a guide to people to get back closer to themselves — not to me.”
Big Thief Tour Dates
Nov. 17 – Tokyo, Japan @ The Garden Hall
Nov. 25 – Melbourne, Australia @ Melbourne Recital Hall
Jan. 31 – Burlington, VT @ Higher Ground
Feb. 3 – New Haven, CT @ College Street Music Hall
Feb. 4 – Philadelphia, PA @ Franklin Music Hall
Feb. 5 – Pittsburgh, PA @ Stage AE
Feb. 7 – Nashville, TN @ The Ryman
Feb. 9 – Oxford, MS @ Lyric Oxford
Feb. 10 – Fayetteville, AR @ George’s Majestic Lounge
Feb. 11 – Tulsa, OK @ Cain’s Ballroom
Feb. 15 – Austin, TX @ ACL at Moody Theatre
Feb. 16 – Houston, TX @ White Oak Music Hall
Feb. 17 – New Orleans, LA @ Orpheum Theater
Feb. 18 – Birmingham, AL @ Iron City
Feb. 20 – Orlando, FL @ Beacham Theatre
Feb. 21 – Miami, FL @ North Beach Bandshell
Feb. 24 – Atlanta, GA @ The Eastern
Feb. 25 – Asheville, NC @ Thomas Wolfe Auditorium
March 2 – New York, NY @ Radio City Music Hall
Apr. 5 – Gateshead, UK @ Sage Gateshead
Apr. 6 – Edinburgh, UK @ Usher Hall
Apr. 7 – Manchester, UK @ O2 Apollo
Apr. 8 – Cardiff, UK @ Great Hall
Apr. 11 – London, UK @ O2 Academy Brixton
Apr. 12 – London, UK @ Eventim Apollo
Apr. 15 – Rotterdam, NL @ Motel Mozaïque Festival
Apr. 16 – Groningen, NL @ SPOT/De Oosterpoort
Apr. 18 – Antwerp, BE @ De Roma
Apr. 22 – Munich, DE @ Muffathalle
Apr. 23 – Milan, IT @ Alcatraz
Apr. 26 – Barcelona, ES @ Sala Razzmatazz
Apr. 27 – Valencia, ES @ Sala Moon
Apr. 28 – Madrid, ES @ La Riviera
Apr. 29 – Lisbon, PT @ LAV Lisboa ao Vivo
We want to hear it. Send us a tip using our anonymous form.
Copyright © 2022 Penske Business Media, LLC. All Rights reserved.