GenomeProt Help Guide

Overview

GenomeProt is a comprehensive proteogenomic analysis tool used to identify:

  • Canonical proteins
  • Noncanonical proteins (uORFs, dORFs, noncoding RNA–derived proteins)
  • Variant peptides and proteoforms

The GenomeProt workflow

The workflow consists of four steps:

  1. Database Generation
  2. Analyse MS Proteomics
  3. Integrate Data
  4. Visualise Results

1. Database Generation

Users can choose between short-read and long-read options depending on their data type.

The public GenomeProt server supports database generation using:

  • Pre-aligned BAM files
  • GTF file of assembled transcripts from the dataset

Users can also generate a variant-aware proteome database using an optional multisample VCF file generated from variant calls derived from the same samples.

Steps to generate a database using the test data

  1. Navigate to the 'Generate database' tab.
  2. Toggle the checkbox 'Incorporate SNVs into protein sequences'.
  3. Unzip and select the corresponding files in the testdata/ directory of the GenomeProt GitHub repository to upload:
    • The reference genome FASTA file (GRCh38_chr1_6_7_masked.fa)
    • The BAM file (long_read_bam/Melanoma_data_subset.bam)
    • The reference transcriptome GTF file (gencode_v47_sorted.gtf)
    • The VCF file (BRAF_mutation.vcf)
  4. Click 'Submit'. The server will then take roughly 1-2 minutes to process the uploaded files.
  5. Once the server has finished processing the files, the 'Download results (zip)' button will be enabled and turn green. Click on it to download a ZIP of the output files.
Database Generation screenshot

Output files:

  • proteome_database.fasta: Multi-FASTA protein database
  • proteome_database_metadata.txt: TSV file containing annotations for candidate protein sequences
  • proteome_database_transcripts.gtf: GTF file with transcript coordinates used to generate the proteome database

The expected contents of the ZIP output file can be downloaded here.

Note: To generate a database directly from BAM files larger than 20 GB, install GenomeProt locally and run its command-line R and Python scripts instead.

2. Analyse MS Proteomics

This module is not present in the public GenomeProt server. Users interested in performing proteomics searches through GenomeProt should install and use GenomeProt locally.

3. Integrate Data

This step maps peptides identified in the proteomics search to spliced transcript coordinates.

  • Files required from Step 1: proteome_database_metadata.txt and proteome_database_transcripts.gtf
  • Files required from the proteomics search: peptide.tsv (discovered peptides) OR report.pr_matrix.tsv (quantified peptides)

Steps to map peptides

This module consists of two parts: Reformatting proteomics results files and performing proteogenomics integration.

Under 'Part 1: Reformat proteomics results files', upload all of the proteomics results files you have obtained from the proteomics search tool you used, then select that tool from the list provided and click 'Submit'.

After the server has finished processing the provided files, it will create a single reformatted peptide results file (peptide_data.tsv). Click on the enabled and green 'Download reformatted results file (peptide_data.tsv)' button to download the reformatted file. With this file, the user can now proceed to the second part ('Part 2: Upload files to integrate').

To perform proteogenomics integration with the test data, upload the following files:

  1. Download the output ZIP file from Step 1 and unzip it.
  2. Navigate to the 'Integrate data' tab.
  3. Under 'Part 2: Upload files to integrate', select the corresponding files to upload:
    • The reformatted proteomics results file (unzip and select the file here).
    • The proteome_database_metadata.txt file from the ZIP file
    • The proteome_database_transcripts.gtf file from the ZIP file
  4. Click 'Submit'. The server will then take a few minutes to process the uploaded files.
  5. Once the server has finished processing the files, the 'Download results (zip)' button will be enabled and turn green. Click on it to download a ZIP of the output files.
Integrate Data screenshot

Output directory contents:

  • summary_report.html: Summary report of mapped peptides, transcripts, and ORFs
  • report_images/: Folder containing PDF versions of graphs from the summary report
  • peptide_info.tsv: Detailed peptide mapping annotations
  • combined_annotations.gtf: GTF file with mapped peptides, ORF annotations and transcript coordinates for visualisation in IsoVis
  • peptides.bed12: BED12 file with mapped peptide coordinates for visualisation in the UCSC Genome Browser
  • transcripts.bed12: BED12 file with transcripts supported by peptide evidence for visualisation in the UCSC Genome Browser
  • ORFs.bed12: BED12 file with ORFs supported by peptide evidence for visualisation in the UCSC Genome Browser

The expected contents of the ZIP output file can be downloaded here.

4. Visualise Results

This step visualises peptides on transcript and gene coordinates using IsoVis.

  • Click 'Upload data' and select combined_annotations.gtf as the 'transcript data' file (max. 3 GB).
  • Optional: select raw transcript counts as the 'transcript counts' file.
  • Optional: select the peptide intensities file from MS analysis (Step 2) as the 'peptide intensities' file. Note that this is NOT the reformatted file from Step 3.
  • Click 'Apply'.
  • Select your gene of interest. Start typing either its gene name or Ensembl ID into the search box, select it from the list of results displayed, then either press enter or click '>'.
  • Use the different visualisation options provided in IsoVis. For example, parts of the visualisation can be toggled, including protein domain labels in the protein diagram.
  • The entire visualisation can be exported as a PNG, JPEG, PDF or SVG, and individual visualisation components can be exported as SVGs.
IsoVis screenshot 1

IsoVis screenshot 2

Visualisation output:

The IsoVis visualisation displays separate tracks for peptides, ORFs and transcripts. Users can limit their analysis to specific peptides, ORFs or transcripts of interest by hiding irrelevant parts of the visualisation and using the peptide and ORF stacks to highlight specific features.

Overlapping ORFs are shown using a hatching pattern. Coding regions are represented as thick dark grey boxes. Peptides uniquely mapped to ORFs, transcripts, and genes are indicated in orange, cyan, and blue, respectively. Multi-mapping peptides are dark grey.

Below is an example exported PNG image from IsoVis:

IsoVis exported PNG example

For additional IsoVis details, refer to the documentation here.