igv-reports

igv-reports - A Python application to generate self-contained HTML reports for variant review and other genomic applications. Reports consist of a table of genomic sites and an embedded IGV genome browser for viewing data for each site. The tool extracts slices of data for each site and embeds the data as blobs in the HTML report file. The report can be opened in a web browser as a static page, with no depenency on the original input files.

Installation

Prerequisites

igv-reports requires Python 3.8 or greater.

Installing igv-reports

pip install igv-reports

igv-reports requires the package pysam version 0.22.0 or greater, which should be installed automatically. However, on OSX this sometimes fails due to missing dependent libraries. This can be fixed following the procedure below, from the pysam docs;
"The recommended way to install pysam is through conda/bioconda. This will install pysam from the bioconda channel and automatically makes sure that dependencies are installed. Also, compilation flags will be set automatically, which will potentially save a lot of trouble on OS X."

conda config --add channels r
conda config --add channels bioconda
conda install pysam

Creating a report

Reports are created with the command line script create_report, or alternatively python igv_reports/report.py. Command line arguments are described below. Although --tracks is optional, a typical report will include at least an alignment track (BAM or CRAM) file from which the variants were called.

Arguments:

• Required

  • sites VCF, BED, MAF, BEDPE, or generic tab delimited file of genomic variant sites. Tabix indexed files are supported and strongly recommended for large files.
  • --fasta Reference fasta file; must be indexed. This argument should be ommited if --genome is used, otherwise it is required.
  • --genome An igv.js genome identifier (e.g. hg38). If supplied fasta, ideogram, and the default annotation track for the specified genome will be used.*

• The arguments begin, end, and sequence are required for a generic tab delimited sites file.

  • --begin INT. Column of start chromosomal position for sites file. Used for generic tab delimited input.
  • --end INT. Column of end chromosomal position for sites. Used for generic tab delimited input.
  • --sequence INT. Column of sequence (chromosome) name.

• Optional coordinate system flag for generic tab delimited sites file only

  • --zero_based Specify that the position in the sites file is 0-based (e.g. UCSC files) rather than 1-based. Default is false.

• Optional

  • --ideogram FILE. Ideogram file in UCSC cytoIdeo format. Useful when fasta is used to specify the reference.
  • --tracks LIST. Space-delimited list of track files, see below for supported formats. If both tracks and track-config are specified tracks will appear first by default.
  • --track-config FILE. File containing array of json configuration objects for igv.js tracks. See the igv.js wiki for more details. This option allows customization of track parameters. When using this option, the track url and indexURL properties should be set to the paths of the respective files.
  • --roi LIST. Space-delimited list of region-of-interest (ROI) files. See igv.js wiki.
  • --template FILE. HTML template file.
  • --output FILE. Output file name; default="igvjs_viewer.html".
  • --info-columns LIST. Space delimited list of info field names to include in the variant table. If sites is a VCF file these are the info ID values. If sites is a tab delimited format these are column names.
  • --info-columns-prefixes LIST. For VCF based reports only. Space delimited list of prefixes of VCF info field IDs to include in the variant table. Any info field with ID starting with one of the listed values will be included.
  • --samples LIST. Space delimited list of sample (i.e. genotypes) names. Used in conjunction with _ --sample-columns_.
  • --sample-columns LIST. Space delimited list of VCF sample FORMAT field names to include in the variant table. If --samples is specified columns will be restricted to those samples, otherwise all samples will be included.
  • --flanking INT. Genomic region to include either side of variant; default=1000.
  • --standalone Embed all JavaScript referenced via ```