FogPharma’s Post

For the Artificial Intelligence final project, Derek Cisneros and I created an algorithm to diagnose COVID-19, the Flu, and the Common Cold. By analyzing a list of symptoms, our model calculates the probability of each disease. While this is a relatively simple application, it serves as a glimpse into the vast potential of AI in healthcare. This inspiration came from the groundbreaking work of AlphaFold. This AI revolutionized the field by predicting the 3D structures of approximately 200 million proteins! Given that protein misfolding is linked to numerous diseases, this achievement is huge! Previously, modeling protein folding was a time-intensive task spanning years. AlphaFold, however, has drastically reduced this duration, significantly accelerating research and development in disease treatment! The development of AlphaFold itself is a testament to the power of interdisciplinary collaboration, bringing together AI engineers, biologists, chemists, and biophysicists. Working on our project with this spirit of collaboration in mind has not only fueled my creativity but also reinforced my determination. It's thrilling to contemplate the future possibilities as developers continue to partner with experts across various fields. These are great times to be a Computer Scientist! #ai #alphafold #machinelearning #biology #biochemistry #computationalchemistry

Last week, I had the privilege of attending "Machine Learning in Chemoinformatics & Drug Discovery." It was fascinating to learn how Machine Learning and Generative AI are revolutionizing the field of bioinformatics and accelerating drug discovery processes like never before. With the power of AI and ML, scientists can now analyze vast amounts of biological data more efficiently, uncovering hidden patterns and insights through Naive Bayes, Random Forest, K Nearest Neighbor, Support Vector Machine, Multilayer perceptron algorithms, PubChem, Padel-Descriptor, and WEKA. This enables faster identification of potential drug targets, Clinical trials, prediction of drug interactions, and optimization of drug design. From virtual screening and molecular modeling to personalized medicine and precision therapeutics, the possibilities are endless! #MachineLearning #GenerativeAI #Bioinformatics #DrugDiscovery #ArtificialIntelligence

Google DeepMind Introduces AlphaFold 3: A Revolutionary AI Model that can Predict the Structure and Interactions of All Life’s Molecules with Unprecedented Accuracy AlphaFold 3 is a state-of-the-art tool from the Google DeepMind and Isomorphic Labs teams in computational biology for predicting the structure and interactions of all life’s molecules. This model employs a revolutionary diffusion-based architecture to enhance predictions’ accuracy significantly beyond existing tools’ capabilities. AlphaFold 3’s methodological advancements allow for comprehensive and precise modeling of biomolecular interactions that were previously unattainable with older computational techniques. The model utilizes a novel approach by integrating a direct diffusion process that predicts raw atom coordinates, bypassing previous models’ limitations requiring detailed, often unavailable, experimental data. This approach has led to remarkable accuracy improvements in predicting the structure of protein complexes and interactions with small molecules and nucleic acids. For example, AlphaFold 3 achieves an interface accuracy of over 90% across various molecular interactions, substantially improving over traditional docking tools and other predictive models. Quick read: https://lnkd.in/geNfT89g Paper: https://lnkd.in/g2iuGEyP Google DeepMind #artificialintelligence #biotech #bioinformatics #gene #ai

Cutting-edge AI meets molecular sciences at NeurIPS 2023! 🌐🔬 Small-Molecule Discovery: # DL, SMILES, and graph methods advance drug discovery. # Attention-based pooling proves effective. # Generative models aid molecule design using learned reactions. Protein Structure: > AlphaFold predictions match with some discrepancies. > pLDDT scores gauge prediction confidence. > Combining AlphaFold2 and cryo-EM enhances accuracy. Top 3 NeurIPS Posters: > OpenProteinSet: Scale training for structural biology. > ProteinInvBench: Benchmarks Protein Inverse Folding. > Protein Design with Guided Discrete Diffusion. https://lnkd.in/ddQC4S6y

MISATO: A Machine Learning Dataset of Protein-Ligand Complexes for Structure-based Drug Discovery In the dynamic field of AI technology, a pressing challenge for the drug discovery

Transformational Implementations of Deep Learning in Regulatory Genomics and Biological Imaging The convergence of advancements in genomics and imaging has ushered in an era of unprecedented data abundance, revolutionizing our understanding of molecular and cellular processes. Yet, traditional analytical methods struggle to grapple with the complexity and scale of these datasets. Enter deep learning, a branch of machine learning poised to unlock hidden insights and facilitate precise predictions by adeptly navigating vast troves of data. This article delves into the transformative applications of deep learning in regulatory genomics and cellular imaging, elucidating its efficacy, operational mechanisms, and potential hurdles. https://is.gd/x1SJJi #AI #artificialintelligence #Biotechnology #genomics #llm #machinelearning

Is biology at an inflection point? Today, at Ginkgo Bioworks, Inc. Ferment 2024 a panel of AI experts discussed biology, AI, data and the acceleration transforming data into knowledge and treatments for patients. Karen Akinsanya, Molly Gibson, Iya Khalil, Debora Marks, and Anna Marie Wagner, eloquently depicted how advances in biological measurement are generating measurements at unprecedented scale (all tissues, all cells) and resolution (spatial metrics to near or actual atomic scale). These are now intersecting with compute power and artificial intelligence capable of actually moving the needle to making biology a quantitative science.

AI is already smarter than humans- it predicted the structure of 200 million proteins in less than a year which could’ve taken us a billion years. In a recent TED talk, Demis Hassabis, co-founder and CEO of Google DeepMind, highlighted the protein folding problem. Over the past 40 years, experimental biologists have pieced together ~150,000 protein structures. It takes one PhD student their whole PhD (4-5 years) to uncover one structure. But there are 200 million proteins known to nature. AlphaFold predicted the structures of all 200 million proteins known to science in less than a year. This might transform medicine discovery forever. What used to take years might take only months to discover. #ai #tech #innovation

🔬🚀 Google DeepMind's AlphaFold 3 Transforms Molecular Interaction Predictions! 💡🔍 Exciting news from Google DeepMind as they unveil AlphaFold 3, a groundbreaking AI system that revolutionizes the accuracy of molecular interaction predictions, paving the way for transformative advancements in drug discovery and bioengineering. 🌟💊 AlphaFold 3's unprecedented precision in modelling protein structures and interactions promises to accelerate scientific research and unlock new insights into complex biological processes. 🧬⚙️ As professionals in the tech and life sciences industries, let's celebrate the potential of AlphaFold 3 to drive innovation and address pressing challenges in healthcare and beyond. 👩🔬🔬 Let's harness the power of AI to unlock the mysteries of molecular biology and usher in a new era of precision medicine and scientific discovery! 🌐🔬 #GoogleDeepMind #AlphaFold3 #AI #Bioengineering #TechInnovation

What will the future of AI-assisted biological discovery look like? As the use of AI in biology and health is gaining momentum, its not hard to imagine a future in which every biological scientist is supported by AI assistants that help synthesise data, design the next experiment, and enrich our mechanistic understanding. This future holds great potential to finally address the many bottlenecks we face in biomedical innovation, and perhaps lead to a world where we are working smarter, not harder in drug discovery. Recently, I had the great opportunity to present some of our recent work in the domain of AI assistants for biological discovery at the Oxford Machine Learning summer school (OxML) in the ML x Bio&Health track - where we covered: - Where can AI make an impact in Drug Discovery and Development? - Computational experimentation as a fast and cost-effective complement to wet-lab research - AI-guided experimental design and optimized exploration - AI-supported synthesis of evidence for causal understanding - In-silico readouts to accelerate and enrich wet-lab experimentation Slides can be found here: https://lnkd.in/dGFRvP5G At the event, I was deeply impressed with the immense knowledge and potential of the summer school attendees that were keen to potentially contribute more to this exciting area in the future. Many thanks for the amazing organising team at AI for Global Goals (Mona Alinejad, D.Phil. (Oxon), Rochelle E., Reza Khorshidi, D.Phil. (Oxon)), Arash Mehrjou for kindly helping with the Q&A, and the team for the outstanding work!