Reference Genomes

When analyzing data generated by high-throughput genome sequencing instruments, one common task is the comparison of those small genome fragments against a reference genome of a known organism. Because multiple groups may study the same organism, a collection of commonly used reference files have been made available for use in high-performance computing on the on-campus SLURM cluster.

iGenomes

The iGenomes collection of reference genomes was developed by Illumina to provide a canonical source of reference genome information. While that project is no longer actively updated, the available reference files have been copied to a folder which is accessible at /shared/biodata/reference/iGenomes/. The structure of those files is described below, and moving forward additional reference genome indices may be generated which follow the same format and structure.

File Structure

All files described below can be found within the directory /shared/biodata/reference/iGenomes/:

• The top-level folder in the iGenomes directory is named for the source organism: e.g. /shared/biodata/reference/iGenomes/Homo_sapiens;

• The second-level folder is named for the source of the annotation used for the reference genome: e.g. /shared/biodata/reference/iGenomes/Homo_sapiens/UCSC;

• The third-level folder is named for the version of the genome: e.g. /shared/biodata/reference/iGenomes/Homo_sapiens/UCSC/hg19;

• The fourth-level folders are Annotation and Sequence;

  • The Annotation folder contains a folder Genes and may also contain SmallRNA and Variation. The exact files provided for each genome may vary, but Genes may contain genes.bed and genes.gtf to describe exon structure of genes, while SmallRNA may contain hairpin.fa and mature.fa to describe miRNAs;

  • The Sequence folder contains a folder WholeGenomeFasta and Chromosomes which contain the FASTA sequence of the reference genome, either in full or broken out by chromosome, respectively;

  • The Sequence folder also contains a set of references which have been pre-compiled for various alignment / analysis algorithms, which may include:

    • BismarkIndex
    • BlastDB
    • Bowtie2Index
    • BowtieIndex
    • BWAIndex
    • MDSBowtieIndex
    • STARIndex: For STAR < 2.7.6a
    • STAR2Index: For STAR ≥ 2.7.6a

Example

Below is an example of the folders available for the UCSC human genome GRCh37/hg19 reference:

├── Homo_sapiens                        # Organism 
│   └── UCSC                            # Source 
│       ├── hg19                        # Version 
│       │   ├── Annotation
│       │   │   ├── Genes               # Gene annotations 
│       │   │   ├── README.txt 
│       │   │   ├── SmallRNA            # Small RNA annotations 
│       │   │   └── Variation           # Variant annotations 
│       │   └── Sequence 
│       │       ├── AbundantSequences	 
│       │       ├── BismarkIndex        # Index for Bismark 
│       │       ├── BlastDB             # Index for BLAST 
│       │       ├── Bowtie2Index        # Index for Bowtie2 
│       │       ├── BowtieIndex         # Index for Bowtie 
│       │       ├── BWAIndex            # Index for BWA 
│       │       ├── Chromosomes         # Genome sequence by chromosome 
│       │       ├── MDSBowtieIndex      # Index for MDS Bowtie 
│       │       ├── STARIndex           # Index for STAR 
│       │       ├── STAR2Index          # Index for STAR2 
│       │       └── WholeGenomeFasta    # Genome sequence 

RNA-Fusion Reference Database

In addition to the iGenomes references described above, the /shared/biodata/ volume also hosts the reference files needed for the nf-core/rnafusion analysis workflow.

The reference database was downloaded as described with version 1.2.0 of nf-core/rnafusion on January 18, 2022 to /shared/biodata/reference/nfcore/rnafusion/1.2.0/2022-01-18.

The content of that folder includes:

├── arriba/                                  # Reference database for the arriba tool 
├── Homo_sapiens.GRCh38_r97.all.fa           # Human reference genome sequence 
├── Homo_sapiens.GRCh38_r97.cdna.all.fa.gz   # Human reference transcript sequence
├── Homo_sapiens.GRCh38_r97.gtf              # Human reference genome annotations
├── pipeline_info/                           # Reference database download report
└── star-fusion/                             # Reference database for the STAR-Fusion tool

Kraken2

Kraken2 is a taxonomic classification tool which is used to identify the microbes present in a complex mixture from whole-genome shotgun sequencing data. As the reference databases needed to run this tool can be quite laborious to build, a public collection of reference databases can be found at:

• /shared/biodata/microbiome/kraken2

Note: This database is automatically added to the path as KRACKEN2_DB_PATH when loading the Kraken2-2.0.7-beta-foss-2016b-Perl-5.28.0 module

Additional databases can be downloaded as-needed from the Langmead Lab Website.

CellRanger

The software produced by 10X Genomics for the analysis of single-cell sequencing data is called CellRanger. A set of reference databases provided for the analysis of 10X single cell data are available at:

• /shared/biodata/ngs/Reference/10X

Note: This database is automatically added to the path as TENX_REFDATA when loading the CellRanger-4.0.0.eb module

SpaceRanger

SpaceRanger is a software suite developed by 10X Genomics (similar to CellRanger) for the analysis of Visium Spatial Gene Expression data. This tool can be used on the rhino/gizmo cluster by loading the SpaceRanger-1.3.0-GCC-10.2.0.eb module. A set of reference databases provided for the analysis of 10X Visium data are available at:

• Human: /shared/biodata/ngs/Reference/10x/refdata-gex-GRCh38-2020-A
• Mouse: /shared/biodata/ngs/Reference/10x/refdata-gex-mm10-2020-A

ANNOVAR

ANNOVAR is a software tool for the annotation of genomic variants. Reference databases for this tool can be found at:

• /shared/biodata/humandb

GTDB-Tk

GTDB-Tk (source) is a software toolkit for assigning objective taxonomic classifications to bacterial and archaeal genomes based on the Genome Database Taxonomy (GTDB). This tool can be used on the rhino/gizmo cluster by loading the GTDBTk-0.1.3-foss-2016b-Python-3.6.7.eb module. Reference databases for this tool can be found at:

• /shared/biodata/humand/release86

AlphaFold

AlphaFold is a powerful tool for predicting protein structures from primary amino acid sequences. This tool can be used on the rhino/gizmo cluster by loading the AlphaFold-2.1.1 module. When using this tool, the environment variable ALPHAFOLD_DATA_DIR is set appropriately to reference the database files available at:

• /shared/biodata/ngs/Reference/protein

CTAT

STAR-Fusion is a component of the Trinity Cancer Transcriptome Analysis Toolkit (CTAT). STAR-Fusion uses the STAR aligner to identify candidate fusion transcripts supported by Illumina reads. STAR-Fusion further processes the output generated by the STAR aligner to map junction reads and spanning reads to a reference annotation set. STAR Fusion uses CTAT genome lib for human fusion transcript detection. Or users can build their own reference library. The CTAT data is from the Trinity Cancer Transcriptome Analysis Toolkit CTAT databases are downloaded from the Broad

• /fh/scratch/app/CTAT

IRIS

IRIS Isoform peptides from RNA splicing for Immunotherapy target Screening. Paper GitHub IRIS reference data was downloaded from IRIS Data

• /fh/scratch/app/IRIS/

Ongoing Support

To maintain the utility of this data resource, the Bioinformatics Core and the Data Core will incrementally add reference genomes for additional organisms and alignment algorithms. If your research group is interested in using an organism or alignment algorithm which is not currently available, please get in touch.