Agent skill

Alphagenome Finetuning

by genomicsxai in genomicsxai/alphagenome-pytorch

Fine-tune or transfer-learn AlphaGenome-PyTorch on custom genomic data — pick a mode (linear probe, LoRA, Locon, full), train on BigWig tracks with agt finetune, use adapters, delta checkpoints…

Apache-2.0Auto-check passedAI & LLM Engineering

Install Alphagenome Finetuning

skills CLI

$ npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install genomicsxai/alphagenome-pytorch alphagenome-finetuning --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/genomicsxai/alphagenome-pytorch.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/alphagenome-finetuning .claude/skills/alphagenome-finetuning && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "alphagenome-finetuning" agent skill from https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning into .claude/skills/alphagenome-finetuning/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "alphagenome-finetuning", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install genomicsxai/alphagenome-pytorch alphagenome-finetuning --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/genomicsxai/alphagenome-pytorch.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/alphagenome-finetuning .agents/skills/alphagenome-finetuning && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "alphagenome-finetuning" agent skill from https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning into .agents/skills/alphagenome-finetuning/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "alphagenome-finetuning", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install genomicsxai/alphagenome-pytorch alphagenome-finetuning --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/genomicsxai/alphagenome-pytorch.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/alphagenome-finetuning .cursor/skills/alphagenome-finetuning && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "alphagenome-finetuning" agent skill from https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning into .cursor/skills/alphagenome-finetuning/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "alphagenome-finetuning", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/genomicsxai/alphagenome-pytorch.git --path .claude/skills/alphagenome-finetuning

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install genomicsxai/alphagenome-pytorch alphagenome-finetuning --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/genomicsxai/alphagenome-pytorch.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/alphagenome-finetuning .gemini/skills/alphagenome-finetuning && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "alphagenome-finetuning" agent skill from https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning into .gemini/skills/alphagenome-finetuning/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "alphagenome-finetuning", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install genomicsxai/alphagenome-pytorch alphagenome-finetuning

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/genomicsxai/alphagenome-pytorch.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/alphagenome-finetuning .github/skills/alphagenome-finetuning && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "alphagenome-finetuning" agent skill from https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning into .github/skills/alphagenome-finetuning/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "alphagenome-finetuning", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install genomicsxai/alphagenome-pytorch alphagenome-finetuning --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/genomicsxai/alphagenome-pytorch.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/alphagenome-finetuning .opencode/skills/alphagenome-finetuning && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "alphagenome-finetuning" agent skill from https://github.com/genomicsxai/alphagenome-pytorch/tree/main/.claude/skills/alphagenome-finetuning into .opencode/skills/alphagenome-finetuning/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "alphagenome-finetuning", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: alphagenome-finetuning
GitHub stars: 162
Token cost: ~1k tokens
SKILL.md length: 390 words
Files: 1
Skills in repo: 2
Repo updated: Sep 15, 2026
First seen: Oct 7, 2026
Licence: Apache-2.0

At a glance

Fine-tune or transfer-learn AlphaGenome-PyTorch on custom genomic data — pick a mode (linear probe, LoRA, Locon, full), train on BigWig tracks with agt finetune, use adapters, delta checkpoints…

ADAPTING/TRAINING the model on new data
Calls python
Not when running predictions with the pretrained model

What it does

Alphagenome Finetuning is an agent skill from genomicsxai/alphagenome-pytorch. Fine-tune or transfer-learn AlphaGenome-PyTorch on custom genomic data — pick a mode (linear probe, LoRA, Locon, full), train on BigWig tracks with agt finetune, use adapters, delta checkpoints, multi-GPU/sequence parallelism, or the Python transfer API. Use when ADAPTING/TRAINING the model on new data, not when running predictions with the pretrained model.

Its SKILL.md is about 1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering, covering Fine-tuning, Deep learning and Bioinformatics. It works with PyTorch and Python. The repository describes itself as: AlphaGenome PyTorch port. The licence is Apache-2.0.

When your agent uses it

ADAPTING/TRAINING the model on new data
Not when running predictions with the pretrained model

Example prompts

“/alphagenome-finetuning”

Requirements

Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 72268c0. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
- python
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Alphagenome Finetuning loads about 1k tokens when it runs. Until then it costs about 96 tokens; SKILL.md has 390 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~96
When it runs · the whole SKILL.md, loaded when a task matches: ~1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from genomicsxai/alphagenome-pytorch at commit 72268c0, republished under its Apache-2.0 licence (© genomicsxai). 390 words, ~1,035 tokens.

Download SKILL.mdSave it as .claude/skills/alphagenome-finetuning/SKILL.md (or your agent's skills folder).

name: alphagenome-finetuning
description: Fine-tune or transfer-learn AlphaGenome-PyTorch on custom genomic data — pick a mode (linear probe, LoRA, Locon, full), train on BigWig tracks with `agt finetune`, use adapters, delta checkpoints, multi-GPU/sequence parallelism, or the Python transfer API. Use when ADAPTING/TRAINING the model on new data, not when running predictions with the pretrained model.

Fine-tuning AlphaGenome-PyTorch

Read docs/finetuning/ for the full guide — it is the source of truth:

docs/finetuning/index.rst — overview and quick start
docs/finetuning/cli.rst — all CLI flags, YAML configs, delta checkpoints, multi-modality, multi-GPU
docs/finetuning/python_api.rst — transfer API, heads, delta weights
docs/finetuning/adapters.rst — linear probing, LoRA, Locon, IA3, merging
docs/finetuning/api_reference.rst — API reference

agt finetune --help is the ground truth for flags.

Quick orientation

Workflow: load trunk → choose transfer mode → add heads for your tracks → train.

Modes (--mode, default lora): linear-probe (heads only, fastest baseline), lora (recommended), locon (adapts Conv1d layers), lora+locon, full, encoder-only. Escalate only if the cheaper mode underfits.

bash

agt finetune --mode lora \
    --genome hg38.fa \
    --modality atac --bigwig data/*.bw \
    --train-bed train.bed --val-bed val.bed \
    --pretrained-weights model.pth

agt finetune and python scripts/finetune.py are the same code path with the same flags — use agt (it ships with the package; scripts/ only exists in a clone). For multi-GPU, torchrun needs a module target: torchrun --nproc_per_node=2 -m alphagenome_pytorch.cli finetune ...

Modalities: rna_seq, atac, dnase, procap, cage (1bp + 128bp); chip_tf, chip_histone (128bp only).

Optional data prep: agt preprocess scale-bigwig --input *.bw --target 100M (depth-normalize) or agt preprocess bigwig-to-mmap (faster training I/O).

Gene-level RNA-seq (both off by default, see docs/finetuning/cli.rst):

--gene-loss-weight 0.1 adds the cross-track gene-LFC loss over gene bodies. Needs --gtf, rna_seq in --modality, and --track-strands.
--gene-expr-eval reports exon-based gene-expression correlations each validation epoch (rna_seq_gene_log_expr_pearson_*). Needs an annotation with exon rows — --gene-expr-annotation, falling back to --gtf.
Both --gtf and --gene-expr-annotation take parquet or GTF/GFF. Prefer parquet (scripts/convert_gtf_to_parquet.py): seconds vs minutes on startup, and one file with exon rows covers both features.

What a run writes: <output-dir>/<run-name> (default finetuning_output/<timestamp>) gets best_model.pth, checkpoint_epoch{N}.pth (every epoch), config.json and the CSV logs. --save-delta is off by default, so a default run produces nothing shareable. --no-full-checkpoint selects deltas-only; --no-save-checkpoints writes no weights at all.

Show full SKILL.md (120 more words)Show less

Loading: full checkpoints and full exports are self-contained (agt predict --checkpoint X); delta checkpoints, exported deltas and adapter bundles also need --model <base weights>. agt info <file> or describe_checkpoint(path) says which you have. See docs/finetuning/checkpoints.rst.

Gotchas:

--resolutions defaults to 1 (1bp only); use --resolutions 128 for chip_tf/chip_histone.
--locon-targets is empty by default and must be set when Locon is enabled (e.g. down_blocks.5, or down_blocks.4,down_blocks.5).
--save-delta works with every mode except full.
Saving/exporting an adapter (--save-delta, export_delta_weights, or agt adapters export) works only from delta weights/checkpoints — merged adapters or a full-model fine-tune have no adapter weights to extract.
There is no --overlap-lowres flag; it is computed as overlap_highres // 128.

For running predictions rather than training, see the alphagenome-predictions skill and docs/alphagenome-usage.md.

Files

Just SKILL.md in .claude/skills/alphagenome-finetuning of genomicsxai/alphagenome-pytorch.

Open the folder on GitHubat commit 72268c0

Compare with similar skills

Alphagenome Finetuning next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Alphagenome Finetuning compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Alphagenome Finetuning this skillgenomicsxai/alphagenome-pytorch	162	—	~1k	Automated safety check: Pass	Apache-2.0	25 days ago
Deep Learningericrisco/rsc-harness	180	—	~3.4k	Automated safety check: Pass	MIT	yesterday
nanoGPT Training GuideOrchestra-Research/AI-Research-SKILLs	13k	2 repos	~1.7k	Automated safety check: Pass	MIT	3 mo ago
OpenVLA-OFT Fine-TuningOrchestra-Research/AI-Research-SKILLs	13k	—	~3.7k	Automated safety check: Pass	MIT	3 mo ago
Pixi Environment Builderxuzhougeng/wisp-science	1k	—	~3.7k	Automated safety check: Pass	AGPL-3.0	yesterday
Benchmark Pyreflyfacebook/pyrefly	7.1k	—	~1.8k	Automated safety check: Pass	MIT	yesterday

Similar skills

Deep Learning
ericrisco/rsc-harness
A skill your agent uses when training or debugging a neural net in PyTorch — the forward/loss/backward/step loop and its silent bugs, mixed precision (AMP), AdamW/LR schedules, DDP/FSDP/ZeRO…
180 GitHub stars~3.4k tokensUpdated yesterday
AI & LLM EngineeringAuto-check passed
nanoGPT Training Guide
Orchestra-Research/AI-Research-SKILLs
Walks through nanoGPT, Karpathy's compact GPT implementation: training on Shakespeare, reproducing GPT-2, fine-tuning GPT-2 checkpoints and training on your own text.
13k GitHub starsUsed in 2 repos~1.7k tokens
AI & LLM EngineeringAuto-check passed
OpenVLA-OFT Fine-Tuning
Orchestra-Research/AI-Research-SKILLs
Fine-tunes and evaluates OpenVLA-OFT and OFT+ robot policies with LoRA and continuous action heads on LIBERO simulation and ALOHA real-robot setups.
13k GitHub stars~3.7k tokensUpdated 3 mo ago
AI & LLM EngineeringAuto-check passed
Pixi Environment Builder
xuzhougeng/wisp-science
A skill your agent uses when creating, migrating, or debugging pixi environments, especially for scientific Python, bioinformatics, single-cell analysis, CUDA/PyTorch, Jupyter/VS Code kernels…
1k GitHub stars~3.7k tokensUpdated yesterday
Research & ScienceAuto-check passed
Benchmark Pyrefly
facebook/pyrefly
Official
Run Pyrefly benchmarks locally via Buck or Cargo, including PyTorch real-world LSP benchmarks.
7.1k GitHub stars~1.8k tokensUpdated yesterday
AI & LLM EngineeringAuto-check passed
Document Public APIs
pytorch/pytorch
Document undocumented public APIs in PyTorch by removing functions from coverageignorefunctions and coverageignoreclasses in docs/source/conf.py, running Sphinx coverage, and adding the appropriate…
104k GitHub stars~4.2k tokensUpdated today
AI & LLM EngineeringAuto-check passed

More from genomicsxai/alphagenome-pytorch

Alphagenome Predictions
genomicsxai/alphagenome-pytorch
Run AlphaGenome-PyTorch to get genomic track predictions — via the agt predict CLI (single locus, BED regions, whole chromosomes, raw FASTA sequences, or per-gene count tables/AnnData), variant…
162 GitHub stars~868 tokensUpdated 25 days ago
Auto-check passed

Works with

Questions about Alphagenome Finetuning

What does Alphagenome Finetuning do?

Fine-tune or transfer-learn AlphaGenome-PyTorch on custom genomic data — pick a mode (linear probe, LoRA, Locon, full), train on BigWig tracks with agt finetune, use adapters, delta checkpoints…. Alphagenome Finetuning is an agent skill from genomicsxai/alphagenome-pytorch. Fine-tune or transfer-learn AlphaGenome-PyTorch on custom genomic data — pick a mode (linear probe, LoRA, Locon, full), train on BigWig tracks with agt finetune, use adapters, delta checkpoints, multi-GPU/sequence parallelism, or the Python transfer API.

When should I use Alphagenome Finetuning?

Alphagenome Finetuning fits situations like: ADAPTING/TRAINING the model on new data; not when running predictions with the pretrained model.

How do I install Alphagenome Finetuning in Claude Code?

Run `npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a claude-code`. Or copy the skill folder (.claude/skills/alphagenome-finetuning in genomicsxai/alphagenome-pytorch) into .claude/skills/alphagenome-finetuning in your project. Claude Code loads it when a task matches its description.

How do I install Alphagenome Finetuning in Codex?

Run `npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a codex`. Or copy the skill folder (.claude/skills/alphagenome-finetuning in genomicsxai/alphagenome-pytorch) into .agents/skills/alphagenome-finetuning in your project. Codex loads it when a task matches its description.

Can I use Alphagenome Finetuning in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add genomicsxai/alphagenome-pytorch --skill alphagenome-finetuning -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/alphagenome-finetuning, .gemini/skills/alphagenome-finetuning, .github/skills/alphagenome-finetuning and .opencode/skills/alphagenome-finetuning in your project.

What does Alphagenome Finetuning need to run?

Going by SKILL.md and its folder, Alphagenome Finetuning needs the command-line tools its instructions call (python). Our summary lists: Python 3.

Does Alphagenome Finetuning access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Alphagenome Finetuning safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Alphagenome Finetuning use?

Alphagenome Finetuning is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Alphagenome Finetuning use?

About 1k tokens (SKILL.md is roughly 4.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Alphagenome Finetuning?

Skills that share tags, products or a category with Alphagenome Finetuning: Deep Learning (ericrisco/rsc-harness, 180 stars), nanoGPT Training Guide (Orchestra-Research/AI-Research-SKILLs, 13k stars), OpenVLA-OFT Fine-Tuning (Orchestra-Research/AI-Research-SKILLs, 13k stars) and Pixi Environment Builder (xuzhougeng/wisp-science, 1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Alphagenome Finetuning?

genomicsxai (a GitHub organization) maintains it in genomicsxai/alphagenome-pytorch, which has 162 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on September 15, 2026.

Source: genomicsxai/alphagenome-pytorch on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Alphagenome Finetuning

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Fine-tuning AlphaGenome-PyTorch

Quick orientation

Files

Compare with similar skills

Similar skills

Deep Learning

nanoGPT Training Guide

OpenVLA-OFT Fine-Tuning

Pixi Environment Builder

Benchmark Pyrefly

Document Public APIs

More from genomicsxai/alphagenome-pytorch

Alphagenome Predictions

Related

Questions about Alphagenome Finetuning