GenBio AI

🚀 Introducing AIDO.StructureDiffusion: our new generative module for high-fidelity molecular design, built for precision, diversity, and control. With AIDO.StructureDiffusion, researchers can generate monomers, complexes, multi-domain proteins, and antibodies, with fine-grained control over structural motifs at the residue level. It enables: → Residue-wise structural conditioning via CAT classes → Target-guided antibody and nanobody design → Immunoglobulin-specific conditioning (CDR loop length, chain locus, developability) → End-to-end generation of single-chain variable fragments (scFvs) 🔬 Trained on AFDB and RCSB PDB datasets, fine-tuned on antibody structures, this model brings advanced controllability to therapeutic protein design. Whether you're designing a novel binder or engineering antibodies, AIDO.StructureDiffusion provides the structure-aware generation capabilities needed to go from idea to candidate faster and more intelligently. Read the blog: 🔗 https://lnkd.in/eSXcjRfp

We're #hiring a new Research Scientist Intern in Abu Dhabi, Abu Dhabi Emirate. Apply today or share this post with your network.

📢 Happening today! GenBio AI Co-Founder & CTO Le Song will be presenting at the MedInvest Biotech & Pharma Investor Conference in Palo Alto, CA. 📍 Session: Decode Biology Holistically, Revolutionize Biomedicine 🗓️ September 24, 2025 – 3:20 PM PT We are proud to contribute to a gathering with 125+ investors, 50+ presenting companies, and leading voices shaping the future of biotech. 🔗 Learn more and register here: https://lnkd.in/ej5F4Z5C

We're #hiring a new Research Scientist Intern in Paris, Île-de-France. Apply today or share this post with your network.

We're #hiring a new Research Scientist Intern in Palo Alto, California. Apply today or share this post with your network.

🚀 AI × Bio updates We’ve been pushing on protein and spatial omics foundation models. Quick roundup: Retrieval-Augmented Protein Language Models for Structure Prediction — ICML Workshops · GenBio/FM4LS. Zero-shot #1 on ProteinGym. Preprint: https://lnkd.in/gQXB476r · Model: https://lnkd.in/gR_GgGrR AIDO.Tissue: Spatial Cell-Guided Pretraining — ICML Workshops · GenBio/FM4LS. Scalable ST pretraining with cell-guided objectives. Preprint: https://lnkd.in/gijgGh6U · Model: https://lnkd.in/gzkuCbSy Multi-Modal Large Language Model Enables Protein Function Prediction — ICML 2025 Workshops · FM4LS. Preprint: https://lnkd.in/gGN7-GiA · Code: https://lnkd.in/g7wRR8xD Protein Inverse Folding From Structure Feedback — NeurIPS 2025 (main). We close the loop with structure-based feedback to improve inverse folding robustness and sample efficiency. Preprint: https://lnkd.in/gnGQvSgE: https://lnkd.in/gQn54dyN ProtGO: Universal Protein Function Prediction Utilizing Multi-Modal Gene Ontology Knowledge — Bioinformatics (Oxford), 2025. A three-step framework that fuses GO knowledge with multi-modal signals for universal function prediction. Paper: https://lnkd.in/gsc44K7V · Code: https://lnkd.in/g3VdYGpX For the last several years we’ve had multiple bio/ML papers at NeurIPS/ICML each year. Honest take: workshops often spark faster, richer exchanges, while Nature family polishing means the system has often iterated by the time a paper appears. Up next: AIDO.DNA v2—likely via tech reports, code, and model releases rather than a traditional paper. Stay tuned (and ping me if you’d like to collaborate). — 🙏 Thanks to the team and collaborators. 📩 DM for details. #AIxBio #ProteinDesign #SpatialTranscriptomics #FoundationModels #RAG #NeurIPS2025 #GenBio #FM4LS #ComputationalBiology #MachineLearning #OpenScience

🚨 New FM4Bio Seminar Series talk at GenBio AI Abhinav Adduri presents state-conditioned perturbation modeling, a framework for predicting how cells respond to interventions while keeping cell state and perturbation effects separate. Why this matters: • Improves generalization across diverse and messy datasets • Makes predictions more interpretable and robust • Supports real-world applications like CRISPR screens, drug response modeling, and experiment design for rare cell populations The approach also highlights the importance of strong baselines, covariate strategies, and biologically meaningful metrics for building reliable models. 📺 Watch the full talk: https://lnkd.in/efXKtyct  📄 Read the paper on bioRxiv: https://lnkd.in/eyaYsBWf  📝 Blog summary: https://lnkd.in/eSevDTZw

⏰ Happening tomorrow! Don’t miss the next Foundation Models for Biology (#FM4Bio) Seminar on September 17 at 9 AM PT. Haotian Cui will present “Large Models for Single-Cell Omics and Drug Discovery: Data, Pretraining, and Closed-Loop Environment.” The talk will cover the development of scGPT, a generative transformer trained on more than 33 million single-cell profiles, and LUMI-lab, a closed-loop self-driving platform for model-guided mRNA delivery. Together, these advances show how large-scale pretraining and closed-loop systems can accelerate discovery in single-cell biology and therapeutic design. 🔗 Last chance to save your spot → https://lnkd.in/enr3Tq5D

We're #hiring a new Research Engineer Intern in Paris, Île-de-France. Apply today or share this post with your network.

We're #hiring a new Research Scientist Intern in Paris, Île-de-France. Apply today or share this post with your network.