skills.rest
cpanse
cpanseCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

genome-reference-build

Build and deploy genome references on FGCZ infrastructure using ezRun conventions.

Updated May 28, 2026
Search Tags:#bioinformatics#ensembl#gtf#cellranger#fgcz#genome reference#ezrun
One-click install
npx skills add https://github.com/cpanse/skills --skill genome-reference-build-cpanse
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: genome-reference-build
Source: https://github.com/cpanse/skills/tree/main/sequencing-pipelines/skills/genome-reference-build
Command: npx skills add https://github.com/cpanse/skills --skill genome-reference-build-cpanse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Adding new species or updated genome assemblies to the FGCZ reference collection requires a precise multi-step process with strict server, naming, and tooling conventions; mistakes cause segfaults, broken indices, or unusable references. ## Core Features & Use Cases - Reference Construction: Downloads GTF and FASTA files from Ensembl, GENCODE, or NCBI and builds the standard ezRun reference directory structure with IGV genomes and annotation tables. - CellRanger Index Generation: Creates 10x Genomics-compatible indices via cellranger mkref for single-cell workflows. - Deployment & Versioning: Copies validated references to /srv/GT/reference/ and commits build scripts to the GitLab reference_files repository. - Use Case: A bioinformatician needs to add the canary (Serinus canaria) Ensembl Release 115 reference so SUSHI apps can use it; the skill provides the exact R script template, server environment setup, and deployment commands. ## Quick Start Ask the agent to build a new genome reference for your organism and Ensembl release following the FGCZ ezRun conventions on fgcz-r-029.

Frequently Asked Questions about genome-reference-build

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a new genome reference for FGCZ infrastructure?▼

Create an R script using ezRun functions buildRefDir and buildIgvGenome with downloaded GTF and FASTA files from Ensembl or GENCODE, then run it on fgcz-r-029. Deploy the output by copying it to /srv/GT/reference/ and commit the script to the GitLab reference_files repository.

How to create a CellRanger index for a custom genome?▼

Run cellranger mkref with the genome.fa and genes.gtf from the built reference, using about 16 threads and 64GB memory. Move the resulting index into the reference's Annotation release directory as genes_10XGEX_Index.

Why does rtracklayer segfault when building references on compute nodes?▼

The segfault occurs because LD_LIBRARY_PATH is not properly inherited on compute nodes like fgcz-c-053, so libR.so is not found. Run builds on fgcz-r-029 and export LD_LIBRARY_PATH pointing to the R 4.5.0 library directory.

Which genome database should I use: Ensembl, GENCODE, or NCBI?▼

GENCODE is preferred for human and mouse due to comprehensive annotations, Ensembl covers most other organisms, and NCBI/RefSeq is the fallback for rare organisms. Species-specific databases like CGD or WBPS serve special cases.

What happens if makeFeatAnnoEnsembl fails during the build?▼

Failure usually means the BioMart dataset name is wrong or the BioMart server is unreachable. The core reference still works without annotation tables, so you can skip this step or verify the dataset name with biomaRt's listDatasets function.

Related Skills

Explore More Education & Research Skills→
LeonChaoXLeonChaoX
783

ensembl-database

Query Ensembl REST API for genes, sequences, variants, and orthologs.

Community
Intermediate
EpiphyticEpiphytic
7

scvi-tools

Applies deep learning methods to analyze and integrate single-cell genomics data.

Official
Advanced
jaechang-hitsjaechang-hits
298

featurecounts-rna-counting

Count aligned RNA-seq reads overlapping gene features in GTF annotations.

Community
Intermediate
mims-harvardmims-harvard
1.6k

tooluniverse-sequence-retrieval

Retrieve and consolidate biological sequences across public databases with cross-references.

Official
Intermediate