ensemblUtils

Herein are a bunch of scripts that I've used over the years to perform various tasks with the Ensembl API.

I'm putting them here as they may be useful to others.

Requirements

All require the Ensembl API is already installed and available via PERL5LIB. See Ensembl for help.

A few perl modules are required in addition to Ensembl:

• Getopt::Long
• Pod::Usage
• File::Basename

More information regarding each script is available with the --help and --man arguments e.g.

perl add_gene_name_column.pl --man

Annotate Data with Ensembl Information

Scenario: you a have a data file with ensembl genes, but you want more information on the genes rather than just their IDs.

The add_gene_name_column.pl script will do this. By default, it will add gene names as an extra column immediately to the right of the gene IDs (in the first column). All other columns are shifted to the right.

perl add_gene_name_column.pl --in data.tsv --species human --desc --out annotated.tsv

This example adds an additional column with the text description of the gene.

Make a More Dynamic Version of Data Files

Tab-delimited files are a perfect way to work or share data. However, they are not the easiest to explore the data unless you are handy in R. You could view them in Excel, but it has well-documented problems when working with gene IDs.

The create_html_table.pl script creates a dynamic, searchable HTML table enriched with jQuery and dataTables goodness, including direct links to ensembl for known gene IDs (first column).

perl create_html_table.pl --in data.tsv --title "My dynamic data" --check-ensembl --out "data.html"

This example enforces a check of the validity of ensembl gene IDs. Any that are not valid are reported and no link to ensembl provided in the final output.

Find motifs in human gene promoters

Provide the script a list of human genes and it will pull back the defined upstream region and find the motif defined within the sequence.

All matches are reported as a BED file which can be used in any genome browser worthy of the name.

perl annotate_promoters.pl --in genes.txt --length 1000 --genome GRCh37 --motif '[AG]CGTG'

Retrieve full-length transcripts

The default downloads from Ensembl (nor BioMart) include the full-length sequence of the transcriptome nor all transcripts. Typically get the coding sequence for protein-coding genes only.

Use this script to retrieve the whole transcriptome for your species of interest.

perl get_full_transcripts.pl --species Arabidopsis_thaliana --out Arabidopsis_transcriptome.fasta

Annotate BED files with gene names

BED files are ubiquitous in genomics, but often are devoid of any information.

Use this script to quickly provide the genes that overlap each entry in the BED file.

perl annotate_bed.pl --in file.bed --species human --genome-build GRCh37 --out annotated.bed

There's more to come...