Sign in

Bo Wang

@bowang87.bsky.social
2.5K followers 45 following 53 posts

Chief AI Officer @ UHN; Assistant Prof. @ U of Toronto; CIFAR AI Chair @ Vector Institute; AI & Biology

PostsRepliesMedia
Bo Wang @bowang87.bsky.social · 17/03/2025
This pivotal work is the result of a collaborative effort led by Micaela E. Consens, with contributions from Cameron Dufault, Michael Wainberg, Duncan Forster, Mehran Karimzadeh, Hani Goodarzi, Fabian J. Theis, Alan Moses. @uhnresearch.bsky.social @vectorinstitute.ai @uoft.bsky.social
110
Bo Wang @bowang87.bsky.social · 17/03/2025
⚡ Strengths, Limitations, & Future Directions: Gain insights into the current capabilities of genomic AI, its limitations, and the promising avenues for future research and application.​
210
Bo Wang @bowang87.bsky.social · 17/03/2025
📊 Comparative Analysis of Models: We delve into the evolution from sequence-to-function models like DeepSEA and Enformer to sequence-to-sequence models such as DNABERT and Evo, highlighting their respective strengths and applications.​
100
Bo Wang @bowang87.bsky.social · 17/03/2025
🚀 Beyond Transformers—Introducing HyenaDNA: Explore innovative architectures like HyenaDNA, which offer efficient long-range genomic sequence modeling at single nucleotide resolution, pushing the boundaries of genomic research.​
100
Bo Wang @bowang87.bsky.social · 17/03/2025
🧠 Transformers in Genomics: Discover how transformer architectures, renowned for their success in natural language processing, are adept at capturing long-range dependencies in genomic data, leading to more accurate models.​
110
Bo Wang @bowang87.bsky.social · 17/03/2025
Key Highlights: 🔬 The Challenges Addressed by gLMs: gLMs tackle the intricate task of interpreting vast genomic sequences, enabling predictions about gene regulation, variant effects, and more.​
100
Bo Wang @bowang87.bsky.social · 17/03/2025
🔥 Unveiling the Future of Genomics with Genome Language Models (gLMs)! 🔥 Our comprehensive review, "Transformers and genome language models," is finally published in Nature Machine Intelligence! ​ Link: nature.com/articles/s42...
nature.com
Transformers and genome language models - Nature Machine Intelligence
Micaela Consens et al. discuss and review the recent rise of transformer-based and large language models in genomics. They also highlight promising directions for genome language models beyond the tra...
1266
Bo Wang @bowang87.bsky.social · 18/02/2025
🙏 A huge team effort behind this work, with special appreciation to BowenLi Lab for driving the project. Kudos to Haotian Cui, Yue Xu, Kuan Pang, Gen Li and Fanglin Gong!
000
Bo Wang @bowang87.bsky.social · 18/02/2025
🌐 Beyond mRNA drugs, LUMI-lab exemplifies a scalable framework for AI-driven molecular discovery, pushing boundaries in material science & drug delivery. 📜 Read the preprint: 🔗 biorxiv.org/content/10.1... 💻 Code available on GitHub: 🔗 github.com/bowenli-lab/...
biorxiv.org
LUMI-lab: a Foundation Model-Driven Autonomous Platform Enabling Discovery of New Ionizable Lipid Designs for mRNA Delivery
The complexity of molecular discovery requires autonomous systems that efficiently explore vast and uncharted chemical spaces. While integrating artificial intelligence (AI) with robotic automation ha...
100
Bo Wang @bowang87.bsky.social · 18/02/2025
🚀 Why it matters? LNPs are the backbone of mRNA therapeutics, yet discovery has been slow due to data scarcity. LUMI-lab shows that AI-powered autonomous labs can accelerate mRNA delivery innovation🚀💡
100
Bo Wang @bowang87.bsky.social · 18/02/2025
- 1,700+ new LNPs synthesized & tested across 10 iterative cycles - Brominated lipids autonomously identified as a novel structural feature that enhances mRNA transfection—an insight previously unrecognized in LNP design - 20.3% in vivo CRISPR gene editing efficiency in lung epithelial cells
100
Bo Wang @bowang87.bsky.social · 18/02/2025
🔥 Key Highlights: - Foundation model trained on 28M molecules using a three-step strategy: - Unsupervised pretraining to capture broad molecular knowledge - Continual pretraining to specialize in lipid-like molecules - Active learning fine-tuning within a closed-loop experimental system
100
Bo Wang @bowang87.bsky.social · 18/02/2025
🔬 What is LUMI-lab? LUMI-lab integrates molecular foundation models with autonomous robotic experiments to efficiently explore new LNPs (lipid nanoparticles, mRNA delivery vehicles) with minimal wet-lab data.
100
Bo Wang @bowang87.bsky.social · 18/02/2025
How can generative AI and Robotics help advance drug discovery? 🚀 Excited to introduce LUMI-lab! A foundation model-driven Self-Driving Lab (SDL) for autonomous ionizable lipid discovery in mRNA delivery 🤖🔍
121
Bo Wang @bowang87.bsky.social · 18/02/2025
🎉 Results speak for themselves: - 63.1% accuracy on ChestAgentBench - State-of-the-art performance on CheXbench - Outperforms both general-purpose and specialized medical models 🙏 Huge shoutout to Adibvafa, Jun, Alif, and Hongwei for their exceptional work on this project!
030
Bo Wang @bowang87.bsky.social · 18/02/2025
📊 Introducing ChestAgentBench: We're also releasing ChestAgentBench, a comprehensive medical agent benchmark built from 675 expert-curated clinical cases, featuring 2,500 complex medical queries across 7 categories. Check it out: huggingface.co/datasets/wan...
130
Bo Wang @bowang87.bsky.social · 18/02/2025
💡 Key Features: - Unified Framework: Seamlessly integrates specialized medical tools with multimodal large language model reasoning. - Dynamic Orchestration: Intelligent tool selection and coordination for complex queries. - Clinical Focus: Designed for real-world medical workflows and deployment.
110
Bo Wang @bowang87.bsky.social · 18/02/2025
🛠️ Integrated Tools: - Visual QA: CheXagent & LLaVA-Med - Segmentation: MedSAM & ChestX-Det - Report Generation: CheXpert Plus - Classification: TorchXRayVision - Grounding: Maira-2 - Synthetic Data: RoentGen
110
Bo Wang @bowang87.bsky.social · 18/02/2025
🎯 Why MedRAX? While specialized AI models excel at specific chest X-ray tasks, they often operate in isolation. Medical professionals need a unified, reliable system that can handle complex queries while maintaining accuracy. MedRAX bridges this gap!
110
Bo Wang @bowang87.bsky.social · 18/02/2025
What is MedRAX? MedRAX is the first versatile AI agent that seamlessly integrates state-of-the-art chest X-ray analysis tools and multimodal large language models into a unified framework, enabling dynamic reasoning for complex medical queries without additional training.
210
Bo Wang @bowang87.bsky.social · 18/02/2025
Agentic AI Meets Medicine!!! 🔬 Excited to announce MedRAX: a groundbreaking Medical Reasoning Agent for Chest X-ray interpretation, now on arXiv! Paper:https://arxiv.org/abs/2502.02673 Code: github.com/bowang-lab/M...
171
Bo Wang @bowang87.bsky.social · 17/02/2025
Huge shoutout to the incredible PHD students Chloe Wang and Haotian Cui for leading this groundbreaking project! 🎉 Massive thanks to our amazing co-authors Andrew, Ronald, and Hani ( @genophoria.bsky.social )from @arcinstitute.org —this work wouldn't have been possible without you! 👏
010
Bo Wang @bowang87.bsky.social · 17/02/2025
📄 Read the preprint: biorxiv.org/content/10.1... 💻 Explore the code/weights: github.com/bowang-lab/s... #SpatialTranscriptomics #SingleCell #AIResearch #MachineLearning #SpatialData
biorxiv.org
scGPT-spatial: Continual Pretraining of Single-Cell Foundation Model for Spatial Transcriptomics
Spatial transcriptomics has emerged as a pivotal technology for profiling gene expression of cells within their spatial context. The rapid growth of publicly available spatial data presents an opportu...
131
Bo Wang @bowang87.bsky.social · 17/02/2025
✨ Multi-Modal & Multi-Slide Integration – Seamless clustering & spatial domain identification across slides and modalities. ✨ Cell-Type Deconvolution & Gene Imputation – Unlocks cross-resolution & cross-modality harmonization with fine-tuned embeddings.
110
Bo Wang @bowang87.bsky.social · 17/02/2025
✨ Revolutionary MoE Decoders – A cutting-edge Mixture of Experts (MoE) architecture for protocol-aware gene expression decoding. ✨ Spatially-Aware Training Strategy – A neighborhood-based masked reconstruction approach to capture complex cell-type colocalization.
120
Bo Wang @bowang87.bsky.social · 17/02/2025
🔥 Why scGPT-spatial? ✨ A Spatial-omic Foundation Model with Continual Pretraining – Built on scGPT’s robust initialization, it unlocks spatial context in tissues. ✨ SpatialHuman30M Dataset – The largest curated dataset: 30M profiles from Visium, Visium HD, Xenium, and MERFISH across 821 slides.
120
Bo Wang @bowang87.bsky.social · 17/02/2025
🧠 What’s the challenge? Spatial transcriptomics is next-level complex—not only must we model single-cell/spot profiles, but we also need to capture intricate spatial relationships while handling diverse sequencing protocols (imaging-based vs. sequencing-based).
120
Bo Wang @bowang87.bsky.social · 17/02/2025
🚀 Introducing scGPT-spatial! 🧬🌍 A game-changing spatial-omic foundation model, built on the powerful scGPT framework with MoE (mixture of experts) and continually pretrained on a massive 30 million spatial single-cell profiles!
1145
Reposted by Bo Wang
Nature Machine Intelligence @natmachintell.nature.com · 29/01/2025
Our Jan issue is live! nature.com/natmachintell with an article (Yejin Choi et al) and N&V commentary (Molly Crockett) on Delphi, designed to investigate AI moral reasoning. Also read about IntegrateAnyOmics by @bowang87.bsky.social, an unsupervised platform to tackle incomplete multi-omics data.
A robot hand trying to play snooker.
092
Bo Wang @bowang87.bsky.social · 22/01/2025
📄 Learn more: BioRxiv: biorxiv.org/content/10.1... Code: github.com/bowang-lab/M... 🎓 Led by the amazing PhD student Navidi Zeinab , co-supervised with Benjamin Haibe-Kains. Big thanks to Jun Ma, Esteban Miglietta, Le Liu, and Anne Carpenter & Beth Cimini for their invaluable contributions!
biorxiv.org
MorphoDiff: Cellular Morphology Painting with Diffusion Models
Understanding cellular responses to external stimuli is critical for parsing biological mechanisms and advancing therapeutic development. High-content image-based assays provide a cost-effective appro...
030
Bo Wang @bowang87.bsky.social · 22/01/2025
3️⃣ Validation: Tested on 3 public Cell Painting datasets, excelling in fidelity, biological interpretability, and visual quality. 🧬 MorphoDiff bridges the gap between transcriptomic-level tools and rich phenotypic insights from high-content microscopy, driving progress in drug discovery.
130
Bo Wang @bowang87.bsky.social · 22/01/2025
✨ Key Contributions: 1️⃣ Perturbation-guided generation: Predicts high-resolution cellular phenotypes in large fields of view. 2️⃣ Generalization: Works with genetic and chemical perturbations, even unseen interventions.
110
Bo Wang @bowang87.bsky.social · 22/01/2025
Our first accepted paper in 2025: 🚀 Introducing MorphoDiff: Our new diffusion-based generative pipeline for high-resolution cellular morphology prediction, guided by perturbation signals, now accepted at ICLR 2025!
192
Bo Wang @bowang87.bsky.social · 20/01/2025
Thrilled that ‘Foundation Models for Biology’ has been named one of the top 7 technologies to watch by @nature.com ! Thanks for highlighting our scGPT work! Also honored to contribute to such a visionary piece. 2025 will be an exciting year for AI & biology! www.nature.com/articles/d41...
nature.com
Self-driving laboratories, advanced immunotherapies and five more technologies to watch in 2025
Sustainability and artificial intelligence dominate our seventh annual round-up of exciting innovations.
161
Bo Wang @bowang87.bsky.social · 10/01/2025
How to build a virtual cell is the “Holy Grail” of the biology! A great read from The Atlantic @theatlantic.com ! Also thanks for highlighting our scGPT, the model that started the new wave of foundation models in biology! www.theatlantic.com/technology/a...
theatlantic.com
A Virtual Cell Is a ‘Holy Grail’ of Science. It’s Getting Closer.
Large language models may unlock a new and valuable type of research.
0114
Bo Wang @bowang87.bsky.social · 17/12/2024
Our Orthrus, one of the first RNA foundation models, is on #NeurIPS2024 !! More details can be seen: x.com/bowang87/sta...
x.com
x.com
170
Reposted by Bo Wang
phil-fradkin.bsky.social @phil-fradkin.bsky.social · 15/12/2024
Excited to be presenting Orthrus with Ruain Shi and Keren Isaev @karini925.bsky.social today! We will be presenting our spotlight at the workshop on AI for new drug modalities #NeurIPS2024 Come chat about a new approach to mRNA representation learning!
1133
Reposted by Bo Wang
Research at UHN @uhnresearch.ca · 10/12/2024
Three UHN research projects have received over $4M through the Ontario Research Fund, part of a $92M investment in Ontario research! Learn how this funding advances: ▶️#MRI tech ▶️#AI for donor lungs ▶️Muscle & joint #health Congrats to the awardees! 🔗 www.uhnresearch.ca/news/investi...
Collage of five smiling male researchers of diverse backgrounds with institutional logos at the bottom.
3113
Reposted by Bo Wang
Research at UHN @uhnresearch.ca · 02/12/2024
AI: The New Virtual Radiologist 🩻 🫁 Researchers at UHN’s Toronto General Hospital Research Institute are using AI to enhance the accuracy and consistency of organ evaluation for lung #transplantation. Read more ➡️ www.uhnresearch.ca/news/ai-new-...
156
Bo Wang @bowang87.bsky.social · 27/11/2024
Thank you to the global community for citing, sharing, and building on this work! 🙏 Read the paper here: nature.com/articles/s41... Stay tuned for MedSAM2, set to be released next month, offering enhanced efficiency and accuracy for medical video and 3D image segmentation.
071
Bo Wang @bowang87.bsky.social · 27/11/2024
The public code repository has received 3000+ stars, and we’re thrilled to see researchers have utilized MedSAM in various ways, such as adaptation, fine-tuning, and development of new models.
130
Bo Wang @bowang87.bsky.social · 27/11/2024
Grateful to the amazing community advancing medical image analysis. This work introduced the first promptable medical image segmentation foundation model, generalizing across diverse medical imaging modalities and a wide range of anatomical structures or pathologies.
120
Bo Wang @bowang87.bsky.social · 27/11/2024
This is an exciting milestone for our MEDSAM! 🎉 Our paper "Segment Anything in Medical Images" has surpassed 1,000 citations!
1193
Bo Wang @bowang87.bsky.social · 26/11/2024
This is amazing article! Thanks @erictopol.bsky.social for highlighting two of our new foundation models for DNA methylation!
0142
Bo Wang @bowang87.bsky.social · 20/11/2024
thanks @haoyin.bsky.social 🙏🙏
151
Bo Wang @bowang87.bsky.social · 18/11/2024
Welcome to the future of epigenetics and AI-powered biomedicine! Let’s push the boundaries of what’s possible. 💡✨
030
Bo Wang @bowang87.bsky.social · 18/11/2024
Special shoutout to Lucas Camillo, the first author of CpGPT, and Albert Ying, the first author of MethylGPT! And many other co-authors including Raghav Sehgal, Steve Horvath, Vadim Gladyshev, Haotian Cui
130
Bo Wang @bowang87.bsky.social · 18/11/2024
Together, MethylGPT and CpGPT redefine what's possible in DNA methylation analysis: - They advance our ability to predict aging and understand disease mechanisms. - They set the stage for more personalized and data-driven healthcare solutions.
120
Bo Wang @bowang87.bsky.social · 18/11/2024
🎯 Complementary Strengths: - CpGPT: Perfect for precise aging analysis and understanding the local genomic context. - MethylGPT: Ideal for clinical applications, large-scale disease prediction, and handling extensive missing data.
120
Bo Wang @bowang87.bsky.social · 18/11/2024
- Clinical Impact: Predicts diseases across 60+ conditions and analyzes the effects of clinical interventions with high precision. - Massive Training Scale: Trained on 226,555 samples (154,063 post-QC), making it one of the largest models for DNA methylation.
120