Find open-source science resources

Multimodal LLM for scientific charts and diagrams understanding/generation

Universal 3D molecular pretraining framework with 209M conformations, scaling to 1.1B parameters (Uni-Mol2) on 800M conformations for molecular property prediction, docking, and quantum chemistry (ICLR 2023, NeurIPS 2024)

End-to-end RNA 3D structure prediction using RNA language model pretrained on 23.7M sequences, outperforming existing methods and human expert groups on RNA-Puzzles and CASP15 (Nature Methods 2024)

Advanced paper search agent powered by large language models, autonomously invoking search tools, reading papers, and selecting references to deliver comprehensive and accurate results for complex scholarly queries (1.5K+ stars, Apache 2.0, 2024)

General purpose tools for high-throughput catalysis.

Self-Referencing Embedded Strings (SELFIES): A 100% robust molecular string representation.

ValidSense is a toolbox for assessing agreement between two quantitative methods or devices measuring the same quantity using the Limits of Agreement (LoA) analysis, also known as the Bland-Altman analysis.

Automated hypothesis testing with agentic sequential falsifications

Bioinspired multi-agent intelligent graph reasoning system that autonomously traverses ontological knowledge graphs to generate, critique, and refine novel research hypotheses, demonstrated on bio-inspired materials discovery with cross-disciplinary connection mining (MIT Lamm Group, 2024)

Systematic medical RAG toolkit for question answering over PubMed, StatPearls, textbooks, and Wikipedia, supporting multiple retrievers, domain LLMs, and follow-up-query workflows for benchmarked clinical/biomedical QA (ACL Findings 2024)

AI-powered tool that automatically converts academic papers (PDF) into presentation slides

Diffusion-based molecular docking achieving SOTA blind docking performance, treating ligand pose prediction as generative diffusion over SE(3), with DiffDock-L update for improved generalization (MIT CSAIL, ICLR 2023)

Unified benchmarking framework for protein representation learning, providing standardized interfaces for pre-training and diverse downstream tasks including structure prediction, fitness prediction, and property prediction across multiple protein datasets and model architectures (ICLR 2024, 273+ stars, MIT License)

Efficient foundation models for PDEs with pretrained transformer-based neural operators and downstream task fine-tuning pipelines, HuggingFace integration for models and datasets (ETH Zurich CAMLab, arXiv 2024)

PyTorch implementation of neural ODEs

LLM-based molecular optimization tool

Vision-language pathology foundation model using contrastive learning on histopathology image-text pairs, enabling zero-shot classification, slide-level retrieval, and multimodal reasoning across diverse cancer types (Mahmood Lab, 494+ stars)

Minimal-modification vision transformer for skillful and reliable medium-range weather forecasting, introducing weather-specific patch embedding, randomized dynamics forecasting over varying time intervals, and pressure-weighted loss; competitive at short range and outperforming prior methods beyond 7 days on WeatherBench 2 with orders-of-magnitude less training data and compute, with favorable scaling in model size and training tokens (MIT License)

Neural optical understanding for academic documents, transforms scientific PDFs to Markdown with mathematical formula support

Generate comprehensive reviews from arXiv papers and convert to blog posts

Microsoft's AI-powered ab initio biomolecular dynamics simulation achieving quantum-mechanical accuracy for proteins with 10,000+ atoms, orders of magnitude faster than DFT using protein fragmentation and ML force fields (Nature 2024)

Equivariant graph attention Transformer (ICLR2023)

Extension of ProteinMPNN for protein sequence design in the context of small-molecule ligands, metal ions, and nucleic acids, enabling binding site engineering and co-factor redesign (Baker Lab)

Physics-AI hybrid modeling for fine-grained weather forecasting (NeurIPS'24)

Geometric deep learning model predicting transcriptional outcomes of novel single- and multi-gene perturbations using gene–gene knowledge graphs, 40% higher precision than prior methods on combinatorial perturbation prediction (Stanford, Nature Biotechnology 2024)

Open-source medical large language model for complex clinical reasoning, extending the o1 long-chain-of-thought paradigm to biomedical question answering and diagnostic inference (FreedomIntelligence, 1.3K+ stars)

A module for solving and visualizing the Schrödinger equation.

Comprehensive toolkit for high-quality PDF content extraction with layout detection, formula recognition, and OCR

Single-cell transformer foundation model pretrained on 104M human transcriptomes via masked gene prediction, enabling transfer learning for cell type classification, gene network analysis, and in silico perturbation with limited labeled data (Nature 2023, V2 2024)

Large-scale biomolecular instruction dataset for chemistry/biology LLMs (ICLR2024)

Large Language Models for automated open-domain scientific hypotheses discovery (ACL 2024, ICML Best Poster)

Universal chart comprehension and reasoning model

Utility that performs integrated analyses of 'gene' data (a set of genes or other genomic features) with 'peak' data (a set of regions, for example ChIP peaks) to identify the genes nearest to each peak, and vice versa.

Transform arXiv research papers into engaging presentations and YouTube-ready videos

Batteries included genomic analysis pipeline for variant and RNA-Seq analysis, structural variant calling, annotation, and prediction.

Materials informatics benchmark

Convert PDF files into editable slides with three lines of code

Structure-aware prefix adaptation for integrating LLMs with knowledge graphs (ACM MM 2024)

Powerful and flexible machine learning platform for drug discovery, providing comprehensive tools for molecular property prediction, generative models, knowledge graph reasoning, and reaction prediction with PyTorch backend (1.5K+ stars)

Resources on ChIP-seq data which include papers, methods, links to software, and analysis.

UNIX-style FASTA manipulation tools.

Biomedical text generation

Usage-Instructions) - A program to visualize reaction networks.

A benchmarking platform for molecular generation models.

Diffusion model for scalable protein structure design with multi-motif scaffolding capabilities, achieving state-of-the-art designability, diversity, and novelty through SE(3)-equivariant attention and massive data augmentation (AlQuraishi Lab, 2024)

Partial-Order Alignment for fast alignment and consensus of multiple homologous sequences.

Pangolin is a deep-learning based method for predicting splice site strengths (for details, see Zeng and Li, Genome Biology 2022). It is available as a command-line tool that can be run on a VCF or CSV file containing variants of interest; Pangolin will predict changes in splice site strength due to each variant, and return a file of the same format. Pangolin's models can also be used with custom sequences.

Short Python script (using Biopython library functions) to extract sequences from a FASTA, QUAL, FASTQ, or SFF file based on the list of IDs given by a column of a tabular file. The output order follows that of the tabular file, and if there are duplicates in the tabular file, there will be duplicates in the output sequence file.

Automated data visualization with minimal code