/PGPT-Pred

https://plabase.cs.uni-tuebingen.de/pb/form.php?var=PGPT-Pred

Primary LanguagePython

PGPT-Pred

The PGPT-Pred tool is part of the web resource PLaBAse (https://plabase.cs.uni-tuebingen.de/pb/plabase.php) and allows annotation of bacterial plant growth-promoting traits (proteins), short "PGPTs" of single genomes, using blastp+hmmer or IMG-KEGG-annoation Mapper against the PGPT ontology.

When applying PGPT-Pred via the PLaBAse always cite the respective reference:

Patz S, Gautam A, Becker M, Ruppel S, Rodríguez-Palenzuela P, Huson DH. PLaBAse: A comprehensive web resource for analyzing the plant growth-promoting potential of plant-associated bacteria. bioRxiv 2021, https://doi.org/10.1101/2021.12.13.472471 (preprint)

Version History and Warranty

  • PLaBAse v.1.01 - Jan 2022
  • Copyright (C) 2020-2022 Sascha Patz.

NOTE:

This web platform and all its services/tools (e.g., PGPT-db, PGPT-Pred and PIFAR-Pred) come with ABSOLUTELY NO WARRANTY!

The code given here is implemented in the backend of PLaBAse and its PGPT-Pred tool!

Here, we we provide only the ontology and annotation exemplarily, for nitrogenase encoding genes and proteins!

The PGPT-Pred tool is only functional with its entire features and complete PGPT ontology when applied via the online platform (see link above)!*

Dependencies:

  1. Interpreters
  1. Tools
  1. Data
  • PGPT-ontology and respective blastp database (via PLaBAse web resource) version 1.01 (Date: Dec2021):
  • An example is currently located in the source and factors folder, but entire ontology will be applicable via API in future (under developement)
  • PGPT-ontology (nitrogenase, blastp+hmmer): factors/PlantGrowthPromotingTraits.csv
  • PGPT-protein blastp database (nitrogenase, blastp+hmmer): factors/PGPT-blastpdb/nitrogenasePGPT
  • PGPT-ontology (nitrogenase, IMG-KEGG-annoation Mapper): src/pgpt_class_kegg.txt

Input formats:

  1. blastp+hmmer: genomic protein sequences in FASTA format (sorted by genomic location)
  2. IMG-KEGG-annoation Mapper: genomic protein KEGG annotations (received by IMG Server, or customer format)

KEGG-Costumer formats:

Accepted file extensions:

  • fasta, fas, fas_aa, faa, txt (genomic proteins, received by NCBI, RAST, IMG, Uniprot, ...)
  • txt (genomic proteins KEGG annotations, as tab separated file)

Main Scripts (Descriptions)

  • pgpt_blhm.py -h
  • pgpt_comp_fun_ascii_v2.py -h

Pipelines applied:

1. Protein/KEGG Annotation/Mapping against PGPT ontology

  1. blastp+hmmer: genomic protein sequences are aligned against proteins associated with the PGPT ontology and respective PFAM domain comparison is achieved by hmmer against the PFAM domains using pgpt_blhm.py
  2. IMG-KEGG-annoation Mapper: KEGG annotations (one per protein only) are mapped against the PGPT ontology, using pgpt_comp_fun_ascii_v2.py -of pfar_kegg

1.1 File Parser

  • if file format is incompatible try to use our parsers: Scripts/python/

2. Pie Chart generation

  1. based on blastp+hmmer results or all blast hits (ignoring pfam comparison) of PGPTs, by applying: Web-Java-PlugIn
  2. based on KEGG-PGPT mapping, by applying: Web-Java-PlugIn

3. Krona Plot generation (kronaScriptPGPT.r)

  1. based on blastp+hmmer results or all blast hits (ignoring pfam comparison) of PGPTs, by applying: R/kronaScriptPGPT.r -m BH
  2. based on IMG-KEGG-PGPT mapping, by applying: R/kronaScriptPGPT.r -m IMK

Results:

  1. Download: Summary file listing all blastp+hmmer or KEGG-mapped hits of PGPTs
  2. Pie Chart: Pie Chart summarizing either all blastp+hmmer hits or all blast hits (ignoring pfam comparison) or KEGG-PGPT hits in a percentage scale on ontology level 2
  3. Krona Plot: Krona Plot giving an hierachical overview of either all blastp+hmmer hits or all blast hits (ignoring pfam comparison) of PGPTs or KEGG-PGPT hits across all hierarchical levels

Runtime estimation:

  • blastp+hmmer: genome of 3000 genes/proteins approx. 30 minutes, genome of 6000 genes/proteins approx. 60 minutes, dependent on server workload
  • IMG-KEGG-annoation Mapper: approx. 15 minutes, dependent on server workload

FURTHER OPTION: Statistical analysis of PGPT-Pred results (raw counts, Scoary output format)

  • see Statistics/stat.r
  • comprises 3 scripts for:
    1. Genome size, gene count and PGPT count analysis
    2. Scoary output-to-heatmap generation
    3. PCA plot computation
  • can be run for all PGPT counts on the entire ontology levels, see internal script documentation
  • Dependencies (libraries): dplyr, gplots, ggplot2, ggpubr, varhandle, RColorBrewer, FactoMineR, factoextra, corrplot, pca3d, vegan, ecodist, ape, cluster, devtools, ggbiplot