Workflows: panoply_ssgsea_workflow

panoply_ssgsea_workflow

Description

This workflow performs single sample Gene Set Enrichment Analysis (ssGSEA) or PTM-Signature Enrichment Analysis (PTM-SEA) on each column of the input data matrix (panoply_ssgsea). The module also generates an interactive report. Optionally, data can be preprocessed with panoply_preprocess_gct to collapse from protein/peptide-level to gene-centric or site-centric. This workflow executes the following modules:

module	description
panoply_preprocess_gct	creates a gene-centric or site-centric GCT file suitable for ssGSEA or PTM-SEA
panoply_ssgsea	performs single sample Gene Set Enrichment Analysis (ssGSEA) or PTM-Signature Enrichment Analysis (PTM-SEA)
panoply_ssgsea_report	creates an interactive R Markdown report of the ssgsea results

panoply_preprocess_gct

This module collapses a feature-level GCT to a gene-centric (for ssGSEA) or site-centric (for PTM-SEA) level GCT. It can optionally be run before panoply_ssgsea module by toggling the preprocess_gct parameter, to collapse a GCT with redundant features into an appropriately-formatted input file.

panoply_ssgsea

This module performs single sample Gene Set Enrichment Analysis (ssGSEA) or PTM-Signature Enrichment Analysis (PTM-SEA) [1] on each column of the input data matrix. This module is based on the implementation available at the ssGSEA2.0 GitHub repository.

This is an updated version of the original ssGSEA [2,3] R-implementation. Depending on the input dataset and chosen database (gene sets or PTM signatures), the software performs either ssGSEA or PTM-SEA, respectively. The Molecular Signatures Database (MSigDB) [4] provides a large collection of curated gene sets. Gene sets are stored as plain text in GMT format. A current version of MSigDB gene set collections can be found in the db/msigdb subfolder. MSigDB gene sets are realeased under Creative Commons Attribution 4.0 International License. The license terms can be found in thedb/msigdb folder.

File formats supported by ssGSEA2.0/PTM-SEA are Gene Cluster Text GCT v1.2 or GCT v1.3 files. Morpheus provides a convenient way to convert your data tables into GCT format.

For more information about the GSEA method and MSigDB please visit http://software.broadinstitute.org/gsea/.

panoply_ssgsea_report

This module creates an R Markdown report to provide a high-level summary of the panoply_ssgsea module results.

The report provides:

• A heatmap depicting the normalized enrichment scores (NES) of gene sets significant in at least one data column. Significance is defined by parameter fdr in section panoply_ssgsea_report of the cfg_yaml file.

• List of parameters used in the panoply_ssgsea module.

Input

Required inputs:

• input_ds: (.gct file) input GCT file

• gene_set_database: (.gmt file) gene set database

• yaml_file: (.yaml file) master-parameters.yaml

• output_prefix: (String) File prefix for output files.

• preprocess_gct: (Boolean) If FALSE panoply_preprocess_gct will be skipped and the GCT file will be used as is (default: FALSE).

Optional inputs:

panoply_preprocess_gct

• acc_type: (String) Type of accession number in 'rid' object in GCT file ("uniprot", "refseq" (default), "symbol").
• id_type: (String) Notation of site-ids: 'sm' - Spectrum Mill (default); 'wg' - Web Gestalt; 'ph' - Philosopher. Only relevant for PTM-SEA.
• id_type_out (String) Type of site id for output: 'uniprot'(default), 'refseq', 'seqwin'. Only relevant for PTM-SEA.
• level (String) Mode of report:
  • 'ssc' - single-site-centric
  • 'gc' - gene-centric (default)
  • 'gcr' - gene-centric-redundant
• loc (Boolean) If TRUE only fully localized sites will be considered (default: TRUE). Localization infromation is expected to be encoded in the site identifier. Respective parsing rules are determined by '--id_type'.
• gene_col: (String) Name of column listing gene names; used for gene centric reports (default: "geneSymbol").
• seqwin_col: (String) "Column containing flanking sequences, separated by '|'. Only relevant for PTM-SEA and if '--id_type_out' = 'seqwin' (default: 'VMsiteFlanks').
• SGT_col: (String) Column used to collpase subgroup-top (SGT) reports (default: "subgroupNum). Only relevant for Spectrum Mill protein reports.
• mod_res: (String) Modified residues, e.g. "S|T|Y" or "K" (default: "S|T|Y").
• mod_type: (String) Type of post-translational modification, e.g "p" for phospho (default) or "ac" for acetylation
• mode: (String) Determines how multiple features (e.g. proteins, PTM sites, etc.) mapping to the same gene symbol will be aggregated:
  • "mean" - mean
  • "median" - median
  • "sd - most variable (standard deviation) across sample columns
  • "SGT" - subgroup top: first subgroup in protein group (Spectrum Mill)
  • "abs.max" - for log-transformed, signed p-values"

panoply_ssgsea

• correl_type: (String) Correlation type: "z.score" (default), "rank", "symm.rank".
• global_fdr: (Boolean) If TRUE global FDR across all data columns is calculated (default: FALSE).
• min_overlap: (Integer) Minimal overlap between signature and data set (default: 10).
• tolerate_min_overlap_err (String) true/false toggle for tolerating "not-enough-overlap" errors. Recommended value is FALSE.
• nperm: (Integer) Number of permutations (default: 1000).
• output_score_type: (String) Score type: "ES" - enrichment score, "NES" - normalized ES (default).
• sample_norm_type: (String) Sample normalization: "rank"(default), "log", "log.rank"
• statistic: (String) Test statistic: "area.under.RES" (default), "Kolmogorov-Smirnov"
• weight: (Float) When weight=0, all genes have the same weight; if weight>0 actual values matter and can change the resulting score (default: 0.75).
• output_prefix: (String, default="results-ssgsea") File prefix for output files.

Output

panoply_ssgsea_workflow produces the follow outputs:

• results: (.tar) An output tar file that contains the ssgsea analysis results
• report: (.html file) Interactive R Markown report.