Computational Biology

Mechanisms of action from drug to disease.

Large-scale biological knowledge graph for drug discovery.

Heterogeneous network integrating genes, diseases, drugs, pathways, and more.

Multi-modal precision medicine knowledge graph integrating clinical, genetic, and drug data.

A collection of samples and tutorials for Microsoft Machine Learning.

Attention-based model for drug response prediction with gene explainability.

GCN + heterogeneous network.

Machine learning framework for predicting synergistic drug combination responses across cell lines.

Multi-view embedding neural network.

GNN embedding + attention mechanism.

Tumor gene set and attention-based model leveraging biological pathway knowledge for drug response prediction.

Hierarchical network model incorporating gene and pathway-level information for cancer drug response prediction.

Deep generative model for predicting transcriptional responses to novel chemical perturbations for drug discovery.

Compositional perturbation autoencoder for predicting single-cell transcriptional responses to unseen drug perturbations and dose combinations.

Interpretable cycle-consistency framework for modeling cellular responses to drug perturbations.

Ensemble machine learning framework combining standard ML with deep learning to systematically rank anti-cancer drugs from proteomics and RNA-seq data.

Deep learning library for drug repurposing.

Dual-VAE architecture for ligand-based virtual screening, drug response prediction, and drug repurposing using chemical-induced transcriptional profiles.

Library for drug-target interaction prediction.

Network-based framework integrating heterogeneous biological data for DTI prediction.

Deep learning model using CNNs on protein sequences and drug SMILES.

Graph neural network–based DTI prediction using molecular graphs.

Transformer-based DTI model leveraging molecular substructures.

Bilinear attention network for interpretable DTI prediction.

Drug discovery via compound-protein interaction and machine learning.

CPI prediction using Transformer.

Reinforcement learning for de novo drug design.

Transformer-based model for molecular generation.

Sequence-to-sequence model for retrosynthesis prediction.

3D equivariant diffusion model for structure-based drug design.

Diffusion generative model for molecular docking, predicting the binding pose of small molecules to protein targets.

Junction tree variational autoencoder for molecular graph generation that guarantees chemical validity via a hierarchical tree decomposition.

Equivariant diffusion model for structure-based drug design that generates molecules and binding conformations for protein targets.

Deep reinforcement learning framework for de novo drug design combining a generative and predictive model.

Reinforcement learning-based generative model for de novo hit-like anticancer molecule design from transcriptomic data.

LLM for biomedical text generation.

LLM for biomedical information, integrated with various APIs.

Foundation LLM for single-cell data.

Pretrained on 50M cells for scRNA-seq denoising & zero imputation.

Bioinformatics-native AI agent skill library with local-first pharmacogenomics, ancestry PCA, semantic similarity, nutrigenomics, and metagenomics skills.

Language model for molecular tasks bridging text and SMILES, enabling molecule captioning and text-driven molecule generation.

LLM-based conversational pipeline for drug discovery, using natural language prompts for iterative drug editing and optimization.

Large-scale foundation model for single-cell gene expression, enabling multiple downstream tasks.

Transformer-based foundation model pretrained on millions of single-cell profiles.

Foundation model for bulk RNA-seq data; learns general transcriptomic representations.

BERT-based foundation model pretrained on large-scale scRNA-seq data for cell type annotation.

Cell pre-trained language model with inter-cell transformer architecture for diverse single-cell analysis tasks.

Universal Cell Embeddings: zero-shot single-cell embedding model trained on 36M cells across species, tissues, and assays without fine-tuning.

Graph-based model for predicting transcriptional responses to single and combinatorial genetic perturbations using biological priors.

Context-aware, attention-based deep learning model pretrained on a large corpus of single-cell transcriptomes.

Transformer-based model integrating gene expression and protein sequences via a protein language model to learn unified multi-species cell embeddings.

Single-cell RNA-seq foundation model trained exclusively on a curated dataset of malignant cells to learn cancer-specific embeddings.

Slide-level digital pathology foundation model pretrained on 1.3 billion pathology image tokens from whole-slide images.

General-purpose self-supervised pathology foundation model trained on 100K+ whole-slide images for diverse computational pathology tasks.

Vision-language foundation model for computational pathology trained with contrastive captioning on pathology image–text pairs.

ViT-based pathology foundation model pretrained with iBOT self-supervision on TCGA whole-slide images.

Foundation model for single-cell and spatial omics using a transformer architecture with positional embeddings to encode spatial cell information.

Extension of scGPT for spatial transcriptomics with continual pretraining and a mixture-of-experts decoder for spatial gene expression analysis.

Single-cell multi-omic language model pretrained on ~10M cells spanning transcriptomics, epigenomics, and proteomics for cross-omics transfer tasks.

Multi-modal variational autoencoder for integrating paired and unpaired single-cell RNA-seq and ATAC-seq measurements into a unified latent space.

Probabilistic multimodal topic model jointly modeling single-cell transcriptomics and chromatin accessibility for regulatory network inference.

Graph-Linked Unified Embedding framework for unpaired single-cell multi-omics data integration across RNA, ATAC, methylation, and protein modalities.

Cross-modality translation model enabling prediction between scRNA-seq and scATAC-seq profiles without requiring paired single-cell measurements.

Asymmetric multi-omics variational autoencoder for integrating single-cell data across RNA, ATAC, and protein modalities with missing-modality support.

Multi-Omics Factor Analysis framework identifying shared axes of variation across bulk and single-cell datasets including RNA, ATAC, proteomics, methylation, and copy number.

Large-scale foundation model integrating DNA regulatory sequences and single-cell transcriptomics from 120M+ cells across multiple species for gene regulation prediction.

Interpretable multi-task deep neural network for single-cell multi-omics integration spanning transcriptomics, chromatin accessibility, and proteomics.

Graph attention network for spatial multi-omics integration jointly embedding spatial transcriptomics with chromatin accessibility or proteomics.

Mosaic integration and differential accessibility model for single-cell multi-omics data that handles arbitrary missing-modality combinations across transcriptomics, chromatin accessibility, and proteomics.

Contrastive self-supervised learning framework for single-cell multimodal data integration, batch correction, and reference-query mapping.

Dual-aligned variational autoencoder for single-cell cross-modality translation between paired and unpaired multiomics data.

Joint variational autoencoder for multimodal single-cell data imputation and embedding.

Bidirectional feedforward network for single-cell multimodal analysis with cross-modality prediction leveraging single-cell atlases.

Transfer learning framework for mapping new single-cell datasets onto pre-trained reference atlases across batches, conditions, and modalities.

Transformer-based framework for one-stop interpretable cell-type annotation supporting cross-dataset and cross-species transfer.

Fast protein structure prediction using language model embeddings.

Predicts structures of proteins, nucleic acids, small molecules, and their complexes.

Open-source all-atom biomolecular structure prediction model for proteins, nucleic acids, small molecules, and their complexes achieving AlphaFold3-level accuracy.

Unified molecular structure prediction model covering proteins, nucleic acids, small molecules, and complexes.

Multimodal protein language model that jointly reasons over sequence, structure, and function for generative protein design and engineering.

Generative model for protein backbone design using diffusion.

Deep learning model for protein sequence design given backbone structure.

High-resolution de novo protein structure prediction from sequence.

Three-track neural network for protein structure prediction.

Trainable, memory-efficient open-source reproduction of AlphaFold2 enabling custom protein structure prediction workflows.

Structure-aware protein language model using structure-aware tokens that encode both sequence and backbone geometry for improved function prediction.

Discrete diffusion framework for protein sequence generation trained on evolutionary-scale data, supporting unconditional generation, disordered region design, and functional motif scaffolding. [paper-2023 ]

RoBERTa-based molecular language model pretrained on SMILES for small-molecule representation learning.

Self-supervised graph transformer for large-scale molecular representation learning from unlabeled compounds.

Unsupervised molecular embedding method inspired by Word2Vec for learning vector representations of chemical substructures.

Linear attention transformer pretrained on millions of SMILES strings for efficient molecular embeddings.

3D molecular pretraining framework for universal representation learning on molecules and protein pockets.

Clinical Histopathology Imaging Evaluation Foundation model integrating histology images and clinical context for pan-cancer analysis.

CLIP-based vision-language foundation model for biomedical images and text trained on PubMed figure–caption pairs.

Pan-cancer integrative histology-genomic analysis framework using multimodal deep learning for patient stratification.

Integrated framework fusing histopathology and genomic features via CNN, GNN, and attention gating for cancer diagnosis and prognosis.

Million-slide digital pathology foundation model using a vision transformer and self-supervised distillation for tile-level pathology image representation.

Tumor Origin Assessment via Deep-learning; weakly-supervised multi-task model predicting cancer primary origin from H&E whole-slide images.

Vision-language foundation model for pathology trained with contrastive learning on pathology image–text pairs for image classification and text-to-image retrieval.

Vision-language foundation model for precision oncology analyzing multimodal paired text and pathology image data for biomarker prediction and retrieval.

Foundation model for genomic sequences across multiple species.

Pre-trained bidirectional encoder for DNA sequence analysis.

Improved genome foundation model with efficient tokenization.

Transformer model predicting gene expression from DNA sequence.

Sequential regulatory activity prediction from DNA sequences.

Bidirectional equivariant long-range DNA sequence model based on Mamba.

Long-context genomic foundation model (up to 1M tokens).

Long-range genomic foundation model handling sequences up to 1M tokens with sub-quadratic attention.

Extended successor to Enformer for predicting RNA-seq coverage from long genomic sequence windows (524 kb) with improved resolution.

Sequence-to-function framework learning a genome-wide regulatory activity code from DNA sequences for variant effect prediction.

Masked language model for DNA sequences enabling zero-shot variant effect prediction without requiring functional annotations.

Suite of protein language models (ProtBERT, ProtT5, ProtXLNet) trained on billions of protein sequences from UniRef and BFD.

Protein language model trained on diverse protein families for sequence generation and fitness prediction.

Efficient protein language model optimized for downstream prediction tasks including secondary structure, localization, and function annotation.