AI

Unified deep learning model that predicts molecular phenotypes from DNA sequence—including gene expression, chromatin accessibility, histone marks, TF binding, splicing, and contact maps—at single-nucleotide resolution for variant effect interpretation.

Deep learning models that predict functional genomic measurements from DNA sequences are powerful tools for deciphering the genetic regulatory code. Existing methods involve a trade-off between input sequence length and prediction resolution, thereby limiting their modality scope and performance1-5. We present AlphaGenome, a unified DNA sequence model, which takes as input 1 Mb of DNA sequence and predicts thousands of functional genomic tracks up to single-base-pair resolution across diverse modalities. The modalities include gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chromatin contact maps, splice site usage and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest available external models in 25 of 26 evaluations of variant effect prediction. The ability of AlphaGenome to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically relevant variants near the TAL1 oncogene6. To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.

A general-purpose biomedical AI agent designed to autonomously execute diverse biomedical tasks across 25 domains. It integrates large language models (LLMs), retrieval-augmented planning, and code-based execution to perform workflows like causal gene prioritization, drug repurposing, and rare disease diagnosis without task-specific tuning.

OpenAI’s consumer and API chat lineup (GPT family), including multimodal and agent-style capabilities.

Anthropic’s Claude family of assistants and API models, emphasizing long context, safety, and agentic workflows.

Anthropic’s terminal-based coding agent for large-scale refactors, test runs, and repo navigation using Claude models.

Autonomous coding agent for VS Code (and compatible editors) that plans, edits files, runs commands, and uses browser tools with user approval.

OpenAI’s cloud and local coding agent (CLI and integrations) for building, reviewing, and shipping code with GPT-based models.

AI-native code editor (VS Code fork) with inline chat, agent mode, and codebase-wide context for editing and refactoring.

DeepSeek’s R1 / V3 and related open-weights and API models focused on reasoning, coding, and efficiency.

A genomic foundation model using the StripedHyena 2 architecture, trained autoregressively on OpenGenome2 (trillions of nucleotides across prokaryotic, eukaryotic, archaeal, and phage genomes) at single-nucleotide resolution with long context (up to about one megabase). Supports generalist prediction and design tasks spanning DNA, RNA, and proteins; code and weights are open source with Hugging Face checkpoints.

All of life encodes information with DNA. Although tools for genome sequencing, synthesis and editing have transformed biological research, we still lack sufficient understanding of the immense complexity encoded by genomes to predict the effects of many classes of genomic changes or to intelligently compose new biological systems. Artificial intelligence models that learn information from genomic sequences across diverse organisms have increasingly advanced prediction and design capabilities1,2. Here we introduce Evo 2, a biological foundation model trained on 9 trillion DNA base pairs from a highly curated genomic atlas spanning all domains of life to have a 1 million token context window with single-nucleotide resolution. Evo 2 learns to accurately predict the functional impacts of genetic variation-from noncoding pathogenic mutations to clinically significant BRCA1 variants-without task-specific fine-tuning. Mechanistic interpretability analyses reveal that Evo 2 learns representations associated with biological features, including exon-intron boundaries, transcription factor binding sites, protein structural elements and prophage genomic regions. The generative abilities of Evo 2 produce mitochondrial, prokaryotic and eukaryotic sequences at genome scale with greater naturalness and coherence than previous methods. Evo 2 also generates experimentally validated chromatin accessibility patterns when guided by predictive models3,4 and inference-time search. We have made Evo 2 fully open, including model parameters, training code5, inference code and the OpenGenome2 dataset, to accelerate the exploration and design of biological complexity.

Google DeepMind’s Gemini model family for chat, search, code, and multimodal tasks across consumer and Vertex / AI Studio APIs.

Zhipu AI’s ChatGLM / GLM family of bilingual LLMs and coding assistants, with open and commercial variants.

xAI’s Grok models integrated with X (Twitter) and standalone apps, aimed at real-time, conversational assistance.

Moonshot AI’s Kimi chat and model lineup, known for long-context reasoning and Chinese–English bilingual use.

Meta’s open-weights Llama family for research and product fine-tuning, from dense LLMs to multimodal stacks.

MiniMax’s text, voice, and video model ecosystem for chat, APIs, and creative / agent applications.

Mistral AI’s open and commercial Mistral / Mixtral models for chat, code, and EU-focused deployments.

A family of transformer foundation models (from tens of millions to multi-billion parameters) pretrained on thousands of human and other-species genomes to learn DNA sequence representations. Embeddings support fine-tuning for tasks such as splice-site prediction, enhancer activity, histone marks, and transcription-factor binding, with benchmarks and weights released openly.

The prediction of molecular phenotypes from DNA sequences remains a longstanding challenge in genomics, often driven by limited annotated data and the inability to transfer learnings between tasks. Here, we present an extensive study of foundation models pre-trained on DNA sequences, named Nucleotide Transformer, ranging from 50 million up to 2.5 billion parameters and integrating information from 3,202 human genomes and 850 genomes from diverse species. These transformer models yield context-specific representations of nucleotide sequences, which allow for accurate predictions even in low-data settings. We show that the developed models can be fine-tuned at low cost to solve a variety of genomics applications. Despite no supervision, the models learned to focus attention on key genomic elements and can be used to improve the prioritization of genetic variants. The training and application of foundational models in genomics provides a widely applicable approach for accurate molecular phenotype prediction from DNA sequence.

Open-source, local-first autonomous AI assistant: file/shell access, browser automation, skills/plugins, and optional chat-app bridges; model-agnostic (e.g. Claude, OpenAI, local/Ollama) with your own API keys.

A deep learning model from Illumina that scores how variants in gene promoter regions alter predicted gene expression, trained on chromatin and expression-related signals at nucleotide resolution. Distributed as a Python package with precomputed genome-wide scores to support rare-disease and research variant interpretation alongside other splice and protein effect tools.

Only a minority of patients with rare genetic diseases are presently diagnosed by exome sequencing, suggesting that additional unrecognized pathogenic variants may reside in noncoding sequence. In this work, we describe PromoterAI, a deep neural network that accurately identifies noncoding promoter variants that dysregulate gene expression. We show that promoter variants with predicted expression-altering consequences produce outlier expression at both the RNA and protein levels in thousands of individuals and that these variants experience strong negative selection in human populations. We observed that clinically relevant genes in patients with rare diseases are enriched for such variants and validated their functional impact through reporter assays. Our estimates suggest that promoter variation accounts for 6% of the genetic burden associated with rare diseases.

Agentic coding platform with IDE, JetBrains plugin, and CLI—quest mode, repo-aware chat, and parallel expert-style agents.

Open-weight and API language / multimodal models (Qwen family) from Alibaba, widely used in research and products.

ByteDance Seed LLM lineup (e.g. Seed 2.0 Pro/Lite/Mini, prior Seed 1.x) for chat, agents, multimodal, and production APIs—sibling to consumer products such as Doubao.

AI-native IDE (ByteDance) with builder mode, multi-model chat, and agent workflows for full-stack development.

Agentic IDE from Codeium with Cascade flow for multi-step edits, terminal integration, and deep workspace awareness.