Meet Evo, an AI model that can predict the effects of gene mutations with ‘unparalleled accuracy’

An image of digitalized DNA with different sequences in bright colours.



The machine learning model Evo can predict and generate sequences of DNA and RNA from their smallest components.
(Image credit: Getty Images/Yuichiro Chino)

Scientists have developed a new type of machine learning model that can understand and design genetic instructions.

The model, dubbed Evo, can predict the effects of genetic mutations and generate new DNA sequences — although those DNA sequences do not closely match the DNA of living organisms.

With time and training, however, Evo and similar models could help scientists understand the functions of various DNA and RNA sequences and mitigate disease, researchers wrote in a new study published Nov. 15 in the journal Science.

Evo is a type of artificial intelligence (AI) system called a large language model (LLM), which is similar to OpenAI’s GPT-4 or Google’s Gemini. Researchers and developers train LLMs on vast amounts of data from publicly available resources, like the internet, and the LLMs look for patterns such as common phrases or typical sentence structures, using those patterns to supply words in a sentence one by one.

Related: Humanity faces a ‘catastrophic’ future if we don’t regulate AI, ‘Godfather of AI’ Yoshua Bengio says

Unlike more common LLMs, Evo isn’t trained on words. Instead, it’s trained on the genomes of millions of microbes — archaea, bacteria and the viruses that infect them, but not eukaryotic organisms like plants and animals. Each base pair — the basic chemical units that make up DNA — from those genomes acts as a “word” in the model. Evo then compares sequences of base pairs against its training set to predict how a strand of DNA will work, or to generate new genetic material.

Other models have already used machine learning and even LLMs to examine genetic information. But so far they have been limited to specialized functions or hampered by high computational cost, the scientists wrote in the study. Evo, by contrast, uses a fast, high-resolution model to process long strings of information, allowing it to analyze patterns at the genome scale and to capture information about large-scale interactions that more specialized models might miss.

Get the world’s most fascinating discoveries delivered straight to your inbox.

The authors tested Evo on a series of tasks. Evo predicted how genetic mutations would affect protein structures, performing comparably to models trained specifically for that task. It also generated one set of protein and RNA components that protected against viral infection in laboratory tests.

Evo even generated sequences of DNA the size of entire genomes — but that DNA wouldn’t necessarily keep something alive. Some of the genetic instructions were similar to DNA in existing organisms. Others looked similar at first glance but didn’t make sense upon closer inspection, similar to an AI-generated image of a person with too many fingers. For example, many of the protein structures encoded in the Evo-generated DNA don’t match naturally occurring proteins.

“These samples represent a ‘blurry image’ of a genome that contains key characteristics but lacks the finer-grained details typical of natural genomes,” the researchers wrote in the study.

They also only trained Evo on microbial genomes, so predicting the effects of human genetic mutations is still out of its grasp. Critically, the team emphasized the need for safety and ethics guidelines to prevent tools like Evo from being misused as their performance improves. In particular, the team excluded data on viral genomes that infect eukaryotic hosts.

“A proactive discussion involving the scientific community, security experts and policy-makers is imperative to prevent misuse and to promote effective strategies for mitigating existing and emerging threats,” the researchers wrote.

Skyler Ware is a freelance science journalist covering chemistry, biology, paleontology and Earth science. She was a 2023 AAAS Mass Media Science and Engineering Fellow at Science News. Her work has also appeared in Science News Explores, ZME Science and Chembites, among others. Skyler has a Ph.D. in chemistry from Caltech.

Note: This article have been indexed to our site. We do not claim legitimacy, ownership or copyright of any of the content above. To see the article at original source Click Here

Related Posts
ANAグループ、就活生向けオンラインイベント 16社参加 thumbnail

ANAグループ、就活生向けオンラインイベント 16社参加

 全日本空輸(ANA/NH)をはじめとするANAグループは、就活生向けのオンラインイベントを1月29日に開催する。グループ16社が参加し、各社の事業や今後の展望などをビデオ会議ツール「Zoom」を通じ紹介する。 就活生向けのオンラインイベントを開くANAグループ=PHOTO: Yusuke KOHASE/Aviation Wire  当日は羽田の空港業務を担うANAエアポートサービスのほか、顧客関連事業や旅行事業などを手掛けるANA Xなど、グループ16社が参加。またANAの人事部が、グループを取り巻く環境についてを講演する。  イベントのエントリーは、ANAグループの採用ページで受け付ける。エントリー締切は1月24日午後11時59分。イベントと採用選考は関係ない。  ANAグループは、グローバルスタッフ職(旧総合職)の2023年度入社の新卒採用を3年ぶりに再開。パイロットや障がい者の採用も実施する。ANA本体など「ANAブランド」の客室乗務員は採用を見送るが、今後はキャリア採用(中途採用)も強化していく(関連記事)。また、グループ約40社の従業員約3万8000人を対象に、働き方の選択肢を広げる制度の導入を検討しており、一部は新年度が始まる4月から開始する(関連記事)。 関連リンクANAグループ採用ページ全日本空輸 ・ANA、グループ内転籍可能に 職種転換も基準緩和(22年1月14日) ・ANA、総合職の新卒採用3年ぶり再開 23年度入社(21年12月27日) ・ANA、退職5年以内の復職制度検討 グループに展開も(21年12月21日) ・ANA、22年3月期最終赤字1000億円に 片野坂HD社長「第4四半期黒字化目指す」(21年10月30日) ・ANA、航空事業2割人員削減 25年度末に3万人体制(21年2月5日) ・ANA、787で中距離国際線LCC新設 アジア豪州方面、エアージャパン母体で22年度就航へ(20年10月27日)
Read More
Women are less likely to seek substance use treatment due to stigma and logistics thumbnail

Women are less likely to seek substance use treatment due to stigma and logistics

According to the Substance Abuse and Mental Health Services Administration, less than 11% of women with a substance use disorder (SUD) received treatment in 2019. Penn State University researchers investigated the barriers women with substance use disorders (SUD) identified that stopped them from seeking treatment.  Abenaa Jones, assistant professor of human development and family studies
Read More
Uranus’ moons could be hiding oceans of water thumbnail

Uranus’ moons could be hiding oceans of water

We may be one step closer to answer one of the many questions surrounding Uranus and its 27 moons. According to a new study, data captured by NASA’s Voyager spacecraft has revealed that four of Uranus’ five major moons could be hiding oceans of liquid water within them. NASA notes that previously it was believed
Read More
Das ist das erste Foto des James-Webb-Teleskops thumbnail

Das ist das erste Foto des James-Webb-Teleskops

© NASA Science 11.02.2022 Das James-Webb-Teleskop hat das erste Foto zur Erde übermittelt, das mit den 18 Spiegeln eingefangen wurde. Die NASA hat das erste Bild des James-Webb-Teleskops veröffentlicht. Zu sehen ist ein Bildmosaik aus 18 anscheinend zufällig angeordneten Sternenlichtpunkten. "Was wie ein einfaches Bild eines verschwommenen Sternenlichts aussieht, bildet nun die Grundlage für die…
Read More
Biologists pinpoint key factor in immune system response to viral infection thumbnail

Biologists pinpoint key factor in immune system response to viral infection

The COVID-19 pandemic has underscored the urgency for science to continue unraveling how viruses infect and how immune systems respond to such threats. University of California San Diego researchers studying how small worms defend themselves against pathogens have discovered a gene that acts as a cell's first-line response against infection. Division of Biological Sciences Postdoctoral…
Read More
Index Of News