AI Revealing Brain Secrets

Ihope you all had a great summer break, at least for those of you in the Northern Hemisphere. During my trip to Singapore and a visit to the family of my best friend, who tragically passed away last year due to a massive burnout, I spent several days reflecting on mental health and our brain.

The brain, with its 100 billion neurons and 200 billion cells, is one of the most complex systems in the universe and holds many mysteries. It is also well recognized that mental health issues are on the rise, and we face a growing number of diseases with unmet needs, such as Alzheimer's.

Attending a meditation class in Bali also raised the question in my mind: how do we measure the benefits on mental health? While our smartwatches provide us with various indicators of our physical health - such as sleep scores, heart rate, and fitness levels - there are few tools that offer insights into our mental well-being.

Could AI help us better understand our brain and pave the way for new treatments?

Mental Health on the Rise. Can AI Help?

There has been a dramatic and alarming increase in mental health issues among young people worldwide - those aged 10 to 24 now account for 45% of the disease burden. The Covid-19 pandemic launched a crisis, triggering anxiety and depression in millions of people. Despite this, only 2% of global health budgets are allocated to mental health care. If more resources are required, how can AI help? New studies show that Large Language Models (LLMs) can identify and predict mental health crises with comparable accuracy to clinicians, but in a significantly shorter amount of time. The findings indicate AI has the potential to support clinicians at a time when there is a severe shortage of behavioral health providers.

Chatbots and facial recognition technology are also increasingly being used to treat and diagnose mental health conditions. As an example, Woebot uses machine learning to understand patients’ messages before delivering pre-written responses created by clinicians. ‘So far, the app—a two-time MedTech Breakthrough award winner—is proving to be popular and convenient. More than 80% [of patients] like using it, a typical interaction lasts just seven minutes, and 77% of interactions occur when providers are off-duty.’

AI-Driven Drug Discovery Advances as MIT Maps Human Brain in 3D

MIT has taken a significant leap in brain research with its innovative technology that maps entire hemispheres of the human brain in 3D. This innovation includes three key technologies: Megatome, which slices intact brain hemispheres without damage; mELAST, a hydrogel that makes brain tissue clear and resilient; and UNSLICE, which reconstructs 3D images from brain slices. The study presents the robust capacities of the groundbreaking technological pipeline. This suite of innovations enables neuroscientists to take a global picture of a hemisphere, zoom in on the subcellular level, map the brain, and understand brain pathology like never before. These advancements allow for detailed imaging of the brain, enabling new insights into brain function, dysfunction and diseases like Alzheimer's. This breakthrough could revolutionize neuroscience by providing unprecedented views of brain pathology.

AI Continues to Revolutionize Drug Discovery

Artificial Intelligence (AI) continues to transform drug discovery, particularly in breast cancer treatment, by accelerating drug development, improving clinical trial matching and enhancing data analysis. For example, AlphaFold, an AI system, has significantly reduced the time needed to solve protein structures, enabling faster development of drugs like HER2-targeted therapies. In another example, AI-powered drug designer Absci has partnered with Memorial Sloan Kettering Cancer Center to co-develop up to six potential antibody therapies in oncology, showcasing AI’s ongoing impact on novel drug design.

Landscape of AI-discovered Drug Candidates

BioPharmaTrend.com highlights the growing landscape of AI-discovered drug candidates, showcasing early successes and the broad potential of AI in pharmaceutical development. These advancements underscore the transformative impact of technology in healthcare, from brain research to personalized medicine.

The chart below highlights the active and competitive field of AI drug discovery, with many companies involved from early discovery to clinical trials including: Insilico Medicine, BenevolentAI, Exscientia, Schrodinger, Recursion Pharmaceuticals, Relay Therapeutics, Valo Health, Verge Genomics, and Insitro. Most candidates are in the early stages, showing the field's youth, but some have reached later clinical phases, indicating early success. AI explores a wide range of drug targets, from established to innovative. Only a few candidates have failed or been discontinued, suggesting promise. The visualization effectively demonstrates AI's potential to accelerate and diversify pharmaceutical development.

This visualization will further evolve with the ISM6331 from Insilico Medicine which has recently received the U.S. Food and Drug Administration (FDA) Investigational New Drug (IND) clearance for the treatment of mesothelioma and the Recursion Pharmaceuticals' recent acquisition of Exscientia. The combined platform of Recursion and Existential now encompasses advanced capabilities such as target discovery, structure-based drug design (including hotspot analysis), quantum mechanics and molecular dynamics modeling. Additional capabilities include: 2D and 3D generative AI design, automated design-make-test-learn cycles with active learning, automated chemical synthesis, predictive ADMET and translation, biomarker selection, and clinical development. This positions Recursion as a leader in the field, strengthened by partnerships with Bristol Myers Squibb (BMS), Sanofi, Bayer, Merck KGaA, Roche/Genentech, and Takeda. This illustration could include many more companies in the future, such as Healx, a U.K. startup using AI to discover new drugs for rare diseases, which has raised $47 million in a Series C round of funding and reportedly received regulatory clearance to start Phase 2 clinical trials for a new drug in the U.S. later this year.

Drug Discovery Deal Making

Those Generative AI platforms drive drug discovery dealmaking. As an example, Insilico Medicine, a pioneer in generative AI for small-molecule drug discovery, recently licensed an IND-ready anticancer inhibitor (USP1) to Exelixis for $80 million upfront. ****A comprehensive table lists these recent partnerships, highlighting the financial terms and collaborative efforts in applying AI to accelerate drug development. Below are some announcements and partnershipsfrom the last few months:

• EvolutionaryScale has introduced ESM3, a groundbreaking generative language model designed to simulate the evolutionary process of proteins by reasoning over their sequence, structure and function. ESM3 marks a significant leap in programming biology, enabling scientists to generate and design new proteins with unprecedented precision, potentially revolutionizing fields such as medicine, biotechnology and environmental sustainability.

• Cognizant partners with NVIDIA to boost drug discovery with AI

• Formation Bio has secured $372m in a Series D financing round to bolster its drug pipeline and expedite the expansion of its AI-driven drug development platform. The company has partnered with OpenAI and Sanofi to design and develop AI solutions for drug development.

• Code Ocean Selected by Lantern Pharma to Power A.I. and Computational Biology Research & Data-Driven Oncology Drug Development Collaborations.

• Israeli startup CytoReason has secured $80 million funding from Nvidia, Pfizer, OurCrowd and Thermo Fisher Scientific to accelerate the development of its AI-powered platform for disease modeling and drug discovery.

• Rejuvenate Biomed and SAS announced today a partnership to develop a user-friendly drug repurposing discovery tool for researchers and scientists. This partnership seeks to uncover medically relevant insights about drugs, diseases and biological networks through fast and systematic explorations of hidden biological patterns.

• Eli Lilly has joined a growing list of pharma companies to partner with OpenAI as it seeks to develop new treatments for drug-resistant pathogens.

Generative AI in Drug Discovery: Synthetic Data, Literature Navigation & Integrated Drug Development Models

Omics synthetic data generation is increasingly crucial in the field of biomedical research because it allows scientists to create artificial datasets that closely mimic real-world biological data. As an example, Synthetic data serves as a valuable resource for developing and fine-tuning computational tools used in genomics. It also allows researchers to simulate various scenarios, making it possible to test hypotheses, refine algorithms and understand the underlying mechanisms of differential expression without needing extensive real-world samples. A new LMM model, Precious2GPT integrates a Multi-omics Pretrained Transformer with Conditional Diffusion to generate synthetic multi-species, multi-tissue omics data, surpassing other models in age prediction accuracy and potential therapeutic target identification. This innovative approach enhances the realism and applicability of synthetic data in clinical and biological research, particularly in aging and cancer studies.

Another challenge that Generative AI is tackling is to leverage efficiently, navigate and utilize the vast body of scientific literature. OpenResearcher is an AI-powered platform designed to tackle the challenge of keeping up with the growing volume of scientific literature. By integrating Retrieval-Augmented Generation (RAG) with Large Language Models (LLMs), OpenResearcher can provide accurate and context-rich answers to research-related queries with up-to-date, domain-specific knowledge. Demo, video, and code are available at: https://github.com/ GAIR-NLP/OpenResearcher.

In the drug discovery space, the newly introduced Tx-LLM stands out as a game-changer. Fine-tuned from the PaLM-2 model, Tx-LLM tackles the complexities of therapeutic development by handling diverse drug discovery tasks using a single model. Its ability to achieve state-of-the-art results across multiple stages of the drug development pipeline, from target discovery to clinical trial predictions, showcases the potential of AI to significantly streamline and accelerate the creation of new therapeutics.

Upcoming 2024 Conferences

• 82nd FIP Wolrd Congress for Pharmacy and Pharmaceutical Sciences, Cape Town September 1st-4th

• BiotechX USA 2024, Philadelphia, PA, September 10-11, 2024

• Intelligent Health AI, Basel, Switzerland, September 11-12, 2024

• Society for Clinical Data Management (SCDM) 2024 Annual conference, Boston,Massachusets, September 29th - October 2nd

• BiotechX Europe 2024, Basel, Switzerland, October 9-10, 2024

• Global health Exhibition, Riyadh, Saudi Arabia, 21st to 23rd October 2024

• Med Tech World, MALTA, November 6th, 8th, 2024

• 6th GCC eHealth Workforce Development Conference, Dubai, November 6th-7th, 2024

• DigiHealthDay 2024, Pfarrkirchen, Germany, November 11th-15th

• GIANT Health 2024 , London, December 9th-10th, 2024

Other Informational Newsletters on Healthcare and AI

Stay up-to-date with the latest insights and industry trends by subscribing to these informative newsletters from my esteemed colleagues and professionals in the field:

• Healthcare in the Metaverse by Philippe GERWILL

• DigitalhealthReadingList by Wolfgang Schleifer

• Future Tech Trends by Bernard Marr

• All about Blockchain Adoption by Kamlesh Nagware

• AI Frontier by Steve Nouri

• The Atlas by Generative AI

• Last Month in Digital health by Gary Monk

AI Basics

DataCamp: AI for Beginners

Google’s AI Video Course

Matt Turck: The AI/ML Landscape

MIT: AI Basics

Nvidia: Jensen Huang on AI

SimpliLearn: AI Bootcamp

WSJ: AI Predictions and Future of Humanity