Score, embed, and generate DNA sequences with the Evo 2 model.
npx skills add https://github.com/guanxiaol/open-science --skill evo2
Please help me install this Agent Skill. Skill: evo2 Source: https://github.com/guanxiaol/open-science/tree/main/runtime/skills/evo2 Command: npx skills add https://github.com/guanxiaol/open-science --skill evo2
The evo2 Skill addresses the challenge of genomic sequence analysis by providing tools for scoring, embedding, and generating DNA sequences, enabling users to work with long-context genomic foundation models efficiently.
Run the evo2 skill to score a DNA sequence using the following command:
python3 score_evo2.py --sequences "ATCG" * 50
High-intent search queries and answers about installing and using this skill.
To score variant effects on a DNA sequence, you can compute per-nucleotide or per-sequence likelihoods using the Evo 2 model. This Skill provides scripts to score regulatory or coding regions across species.
Yes, you can generate DNA sequences conditioned on a specific prefix using the Evo 2 model. This allows researchers to synthesize long-context genomic sequences based on provided starting nucleotides.
You can embed genomic windows for downstream classification tasks by applying the sequence embedding functionality of the Evo 2 model. This processes long-context genomic sequences into numerical representations.
Yes, performing genomic sequence analysis and generation with the evo2 Skill requires a GPU. It also requires specific Python libraries and the Evo 2 dependency to execute long-context genomic foundation model operations.
The best way to analyze long-context genomic sequences for likelihood scoring is using a dedicated genomic foundation model like Evo 2. It enables per-sequence scoring to evaluate variant effects across regulatory and coding regions.
Limitations of using Evo 2 for genomic sequence generation include its strict requirement for a GPU and specific Python libraries. Users must also provide a prefix to condition the DNA sequence generation properly.
PKU-YuanGroupScore, embed, and generate DNA sequences with a genomic foundation model.
PKU-YuanGroupPredict RNA-seq, CAGE, DNase, and ChIP tracks from DNA sequences.
jaechang-hitsTrain and apply region2vec embeddings for genomic interval machine learning.
ovachieverProvides Python-based bioinformatics tools for analyzing DNA, RNA, and protein sequences.