Medical AI could be ‘dangerous’ for poorer nations, WHO warns

The rapid growth of generative AI in health care has prompted the agency to set out guidelines for ethical use...

"The guidelines recommend that independent third parties perform and publish mandatory post-release audits of LMMs that are deployed on a large scale. Such audits should assess how well a tool protects both data and human rights, the WHO adds.

It also suggests that software developers and programmers who work on LMMs that could be used in health care or scientific research should receive the same kinds of ethics training as medics. And it says governments could require developers to register early algorithms, to encourage the publication of negative results and prevent publication bias and hype..."

https://www.nature.com/articles/d41586-024-00161-1

AlphaFold predicts psychedelics

"The protein-structure-prediction AI AlphaFold has helped to identify thousands of potential psychedelic molecules, which could help to develop new antidepressants. This is the first time predicted protein structures have been shown to be just as useful for drug discovery as experimentally derived ones, which can take months to determine. In one out of three cases, an AlphaFold prediction could jump-start a drug discovery project by a couple of years, estimates pharmaceutical chemist and study co-author Brian Shoichet. Others caution that AI-generated molecules seem to be helpful for some biological targets but not others, and it’s not always clear which applies..."

https://www.nature.com/articles/d41586-024-00130-8

I agree that AI has the potential to be the most useful tool in the search for novel targeted treatment drugs, but I also see that there are a lot of reasons why it can fail on occasion. 1) Proteins are continually changing all the time. 2) Pharmacological off-target effects are difficult to detect. Because of its renown for obscurity, AI approaches like deep learning make it hard to discern if they are data-driven solutions or grounded on beneficial criteria for medication determination. The representation of generic data on comorbidities, race, gender, ethnicity, environmental variables, etc. is another major limitation at the moment.

Я в цьому не досвідчена. з повагою Чорна Валентина

Please read the following paper

Artificial intelligence in drug discovery and development

doi: 10.1016/j.drudis.2020.10.010

PMCID: PMC7577280PMID: 33099022

Best of Luck

Reality check for AI in drug discovery

"In a first-of-its-kind test of AI’s ability to come up with promising starting points for new medicines, 23 teams participated in a drug-discovery competition, predicting over 2,000 compounds that could bind to an enzyme associated with Parkinson’s disease. Fewer than a dozen actually bound to the enzyme when tested in the lab. “We're at less than a 1% hit rate,” says computational chemist and competition coordinator Matthieu Schapira. Several top methods used machine learning, but classical methods such as ultra-high throughput docking still kept up. “The question is, do we need really complicated, advanced AI technologies at every stage of the process?” asks pharmaceutical researcher Lukas Friedrich, who was a top-scoring competitor in the challenge..."

https://www.nature.com/articles/d41573-024-00036-0

Safety guidelines for AI-designed proteins

"Researchers have launched an initiative calling for the safe and ethical use of protein design. The voluntary effort comes on the heels of reports from lawmakers, think tanks and other organizations exploring the possibility that AI tools — ranging from protein-structure prediction networks such as AlphaFold to large language models such as the one that powers ChatGPT — could make it easier to develop biological weapons, including new toxins or highly transmissible viruses. The new initiative calls for the biodesign community to police itself and for improved screening of DNA synthesis, a key step in making AI-designed proteins, for potentially harmful molecules..."

https://www.nature.com/articles/d41586-024-00699-0

AI designs drug in just 18 months

"A drug against idiopathic pulmonary fibrosis, created from scratch by AI systems, has entered clinical trials. Researchers at Insilico Medicine identified a target enzyme using an AI system trained on patients’ biomolecular data and scientific literature text. They then used a different algorithm to suggest a molecule that would block this enzyme. After some tweaks and laboratory tests, researchers had a drug that appeared to reduce inflammation and lung scarring. Medicinal chemist Timothy Cernak says he was initially cautious about the results because there’s a lot of hype about AI-powered drug discovery. “I think Insilico’s been involved in hyping that, but I think they built something really robust here.”..."

https://cen.acs.org/physical-chemistry/computational-chemistry/Insilico-reveals-soup-to-nuts-process-for-AI-generated-lung-fibrosis-drug/102/web/2024/03

AI helps design antibodies from scratch

"For the first time, an artificial intelligence (AI) system has helped researchers to design completely new antibodies. Creating new versions of these immune proteins, which can be used as drugs, is usually a lengthy and costly process..."

https://www.nature.com/articles/d41586-024-00846-7

AI can help to tailor drugs for Africa

"Artificial intelligence (AI) systems can help researchers to understand how genetic differences affect people’s responses to drugs. Yet most genetic and clinical data comes from the global north, which can put the health of Africans at risk, writes a group of drug-discovery researchers. They suggest that AI models trained on huge amounts of data can be fine-tuned with information specific to African populations — an approach called transfer learning. The important thing is that scientists in Africa lead the way on these efforts, the group says..."

https://www.nature.com/articles/d41586-024-01001-y

Marc Tessier-Lavigne on leaving Stanford and joining biotech’s new AI mega-startup

"Former Stanford president Marc Tessier-Lavigne will lead one of biotech’s biggest-ever startup launches: Xaira Therapeutics, which has secured over $1 billion to transform drug discovery and development with AI...

The move is sure to raise eyebrows..."

https://endpts.com/marc-tessier-lavigne-on-leaving-stanford-and-joining-ai-mega-startup-xaira/

Major AlphaFold upgrade

"A new version of DeepMind’s AlphaFold tool gives scientists the ability to predict protein structures during interactions with other molecules. The tool could be transformative for drug discovery because it can predict the shape of proteins that contain function-altering modifications, or their structure alongside those of DNA, RNA and other cellular players that are crucial to a protein’s duties..."

https://www.nature.com/articles/d41586-024-01383-z

Podcast: Major AlphaFold upgrade

"“The real revolution was AlphaFold2,” Nature reporter Ewen Callaway tells the Nature Podcast. But “it just predicted protein structures, it didn’t predict proteins alongside all the other players of the cellular ecosystem”, such as DNA. “AlphaFold3 is exactly that.” AlphaFold3 is also reportedly easier to use and incredibly fast. Access to the tool, however, is limited — partly to protect the advantage of DeepMind’s own drug-discovery spin-off company..."

https://www.nature.com/articles/d41586-024-01385-x

Nature earns ire over lack of code availability for Google DeepMind protein folding paper

"A group of researchers is taking Nature to task for publishing a paper earlier this month about Google DeepMind’s protein folding prediction program without requiring the authors publish the code behind the work.

Roland Dunbrack, of Fox Chase Cancer Center in Philadelphia, peer-reviewed the paper but was “not given access to code during the review,” the authors of a letter submitted today, May 14, to Nature – including Dunbrack – write, “despite repeated requests.”..."

https://retractionwatch.com/2024/05/14/nature-earns-ire-over-lack-of-code-availability-for-google-deepmind-protein-folding-paper/

Reality check for AI-designed drugs

"An analysis claiming that AI-discovered drugs were highly successful suffers from small sample size and a vague definition of ‘AI-discovered’, argues medicinal chemist Derek Lowe. The study finds an 80—90% success rate in early clinical trials, compared with a 40—65% historical industry average. Lowe points out that for most of the 24 analysed therapies, the biological target was already known for the disease under investigation. “For now, I am not convinced that issuing press releases about your compounds that talk about their discovery through AI techniques is sufficient to expect greater things from them.”..."

https://www.science.org/content/blog-post/ai-drugs-so-far

Exploring New Frontiers in Pharma: Mindsets, Data, AI, and Oracle

"How can shifting mindsets fuel the next wave of innovation in the pharmaceutical and life sciences industry? In what ways can we ensure the vast amounts of health data are utilized securely and effectively to foster groundbreaking medical advancements? And how is Oracle's new Health Data Intelligence poised to transform the industry in an unprecedented manner?..."

https://researchinaction.libsyn.com/site/exploring-new-frontiers-in-pharma-mindsets-data-ai-and-oracle

Who will make AlphaFold3 open source?

"Researchers are racing to create their own versions of DeepMind’s revolutionary protein-structure-prediction AI AlphaFold3, which was released without its code, to the frustration of many scientists. “It would be bad if capabilities that are just so fundamental to our ability to do drug discovery… end up getting locked up,” says computational biologist Mohammed AlQuraishi, whose team hopes to complete an ‘OpenFold3’ this year. Computational biophysicist David Baker’s team is adapting a version of their open-fold algorithm RoseTTAFold, and independent software engineer Phil Wang has begun a crowdsourced replica of AlphaFold3. Responding to the backlash, DeepMind has promised to release the code by year’s end..."

https://www.nature.com/articles/d41586-024-01555-x

Why AlphaFold 3 Needs to Be Open Source

The powerful AI-driven software from DeepMind was released without making its code openly available to scientists...

"IMAGINE A WORLD where in a matter of minutes, scientists could identify drugs to treat incurable diseases, design chemicals that could break down plastics to clean up pollution, and develop new materials that can suck excess carbon dioxide out of the air to help address climate change. This is the promise of new biology- and chemistry-based models that use artificial intelligence, or AI, to perform traditionally time-consuming tasks such as determining the structures of proteins.

Google DeepMind, a private research subsidiary of Google, released the highly anticipated AlphaFold 3 model last month as a paper in Nature. This model claims to be an improvement over its earlier version, AlphaFold 2, because it can predict not just protein structures, but also how they interact with RNA, DNA, and — most importantly — drugs. DeepMind said that it hopes AlphaFold 3 will “transform our understanding of the biological world and drug discovery.”...

In an editorial response published on May 22, Nature claimed that by allowing peer-reviewed publications from the private sector, it “promotes the sharing of knowledge, verification of the research and the reproducibility researchers strive for” and that its policy states that the editors reserve the right to decide if all code needs to be released. However, it’s unclear to me how one can verify research without having the tools available to do so..."

https://undark.org/2024/06/06/opinion-alphafold-3-open-source/

AI predicts almost 1 million antibiotics

"A machine-learning algorithm sifted through tens of thousands of microbial genomes and predicted more than 800,000 DNA snippets that encode for antimicrobial compounds. More than 90% of these potential new antibiotics had never been described before. Out of the 100 of these compounds the researchers synthesised, three were highly effective in mice against an infection of an often-resistant pathogen..."

https://www.dw.com/en/are-scientists-finally-beating-antimicrobial-resistance/a-69275675

"EvolutionaryScale’s AI tool, called ESM3, is what’s known as a protein language model. It was trained on more than 2.7 billion protein sequences and structures, as well as information about these proteins’ functions. The model can be used to create proteins to specifications provided by users, akin to the text spit out by chatbots such as ChatGPT..."

https://www.nature.com/articles/d41586-024-02214-x