gtars

High-performance toolkit for genomic interval analysis in Rust with Python bindings. Use when working with genomic regions, BED files, coverage tracks, overlap detection, tokenization for ML models, or fragment analysis in computational genomics and machine learning applications.

1.23x

Quality

66%

Does it follow best practices?

Impact

32%

1.23x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/gtars/SKILL.md

Evaluation results

28%

14%

ChIP-seq Peak Overlap Analysis with Regulatory Elements

IGD overlap detection and RegionSet operations

Criteria

Without context

With context

Uses gtars import

50%

IGD index build

IGD query method

IGD index persistence

RegionSet.from_bed usage

25%

filter_overlapping method

Set operation usage

to_bed export

100%

Overlap statistics

CLI IGD commands

Results documented

100%

No large file warning

100%

32%

Single-Cell ATAC-seq Coverage Analysis and Cluster-Based Fragment Processing

Coverage track generation and fragment processing

Criteria

Without context

With context

Uses gtars/gtars CLI

57%

coverage_from_bed or gtars uniwig

BigWig format output

Resolution parameter

Fragment file format awareness

100%

80%

fragsplit filter command

fragsplit cluster-split command

Correct CLI argument syntax

Coverage output file produced

50%

75%

Workflow documentation

100%

85%

No large files

100%

38%

Genomic Region Tokenization for Deep Learning Model Training

Genomic tokenization for ML preprocessing

Criteria

Without context

With context

TreeTokenizer import path

from_bed_file constructor

tokenize method signature

Batch tokenization loop

100%

Token ID access

vocab_size usage

100%

NumPy array export

RegionSet from_bed usage

25%

geniml integration reference

Python API for pipeline

100%

Token output file

100%

No large files

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: b58ad7e

Evaluated: 27 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

ChIP-seq Peak Overlap Analysis with Regulatory Elements Single-Cell ATAC-seq Coverage Analysis and Cluster-Based Fragment Processing Genomic Region Tokenization for Deep Learning Model Training

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.