Skip to content

celawson87/reverseEcology

Repository files navigation

From Genomes to Traits: Reverse Ecology of Uncultivated Freshwater Actinobacteria

Copyright (c) 2015, Joshua J Hamilton and Katherine D McMahon

Department of Bacteriology

University of Wisconsin-Madison, Madison, Wisconsin, USA

All rights reserved.


Data Folders

refGenomes

Reference genomes used in this study. One folder per genome format.

RawModelFiles

Contains raw output from KBase. One folder per genome. To construct a complete and valid SBML file from models built using the Narrative interface, three files are required:

  • AAA027E14.xml - SBML reconstruction from KBase
  • AAA027E14Compounds.tsv - Text format reconstruction from KBase
  • AAA027E14Reactions.tsv - Text format reconstruction from KBase

excludedSamples

Raw and processed model files for genomes which have been excluded from this analysis.

ExternalData

Additional files which are required to make various steps of the pipeline run.

  • For reverseEcology.ipynb:

    • reactionsToRemove.txt - optional file. If you want to remove reactions from the models, list the reactions here.
    • taxonomy.csv - comma-separated file containing the following info for each sample: (ID, Lineage, Clade, Tribe, Color). Should be sorted alphabetically.
    • taxonomyTribes.csv - same as taxonomy.csv, but for all samples which are part of a tribe
  • For pairwiseANI.ipynb:

    • ANI_out - table of pairwise ANI calculations
    • COV_out - table of pairwise coverage calculations
    • taxonomySAG.csv - same as taxonomy.csv, for SAGs only
  • For mergingGenomes.ipynb:

    • taxonomySAG.csv - same as taxonomy.csv, for SAGs only
    • tribalColors.csv - sams as taxonomy.csv, but for tribes
  • Also contains information about the genomes:

    • cogAbundance.csv - number of genes in each genome in each COG. Exported from IMG.
    • checkMMarkerGenes.csv - presence/absence of marker genes used by CheckM
    • phylophlanMarkerGenes.csv - presence/absence of marker genes used by Phylophlan
    • phylosiftMarkerGenes.csv - presence/absence of marker genes used by Phylosift
    • masterColors.xlsx - spreadsheet showing coloration used in figures

metadata

  • 2015-08-21-FinalTree - Phylogenetic tree of genomes with final classifications 2015-08-18-actinoAlignment.phylip - Phylosift alignment used to generate the above tree 2015-08-18-actinoTreeCleanNames.nwk - RAxML-build tree file which I made pretty
  • 2015-08-21-actinosMetadata(formatted).xlsx - metadata about the genomes, including estimated completeness, genome size, etc

Code Folders

Matlab

Codes for processing initial reconstructions from KBase. Deprecated in favor of Python codes.

Perl

Codes for pushing genomes to KBase for model reconstruction. Deprecated since API access to KBase is broken (2016-01-25).

Python

Codes for performing reverse ecology analysis.

  • Function files called by workflow notebooks

    • metadataFunctions.py - functions for importing models and taxonomies
    • markerGeneFunctions.py - functions for computing completeness estimates from presence/absence of marker genes
    • sbmlFunctions.py - functions for working with SBML reconstructions from KBase
    • graphFunctions.py - functions for working with graph representations of reconstructions
    • seedFunctions.py - functions for computing reverse ecology metrics from seed sets
    • pairwiseANIFunctions.py - functions for performing ANI and coverage calculations
    • extraFunctions.py - extra functions which are currently unused in the final analysis
  • Other Files

    • reverseEcology.py - master workflow for performing reverse ecology analysis
    • mergingGraphs.py - master workflow for merging genome network graphs and performing reverse ecology analysis.

Results Folders

DataSummaries

Results of reverse ecology computations. Also contains a folder MergedData which contains results for analyses performed on tribe-level aggregates of genomes.

  • ModelStatistics.txt - Numbers of genes, reactions, and metabolites in the SBML models.
  • seedMatrixWeighted.csv - For each metabolite in the set of models, whether or not the compound is a seed (1 = yes). A metabolite may have value <1 if it can substitute for another compound.
  • SeedCounts.txt - For ALL genomes, frequency with which a metabolite is a seed compound (1 = metabolite is a seed in 100% of genomes)
  • MetabolicCompetitionScores.csv - metabolic competition between pairs of genomes
  • MetabolicComplementarityScores.csv - metabolic complementarity between pairs of genomes
  • metabolicCompetition.png - heatmap of competition scores
  • metabolicComplementarity.png - heatmap of complementarity scores

iPython Notebook

iPython notebook summaries of the project

  • masterNotebook - master workflow for computing reverse ecology metrics.

  • ReverseEcologyOLD - old version of the above.

  • genomesFromKBase.ipynb - workflow for converting SBML models from KBase into a form suitable for work with reverse ecology

  • pairwiseANI - ANI- and coverage-based criteria for identifying tribes. Analysis of which partially-classified GFMs can be merged into tribes.

  • mergingGenomes - workflow for for merging network graphs of individual genomes into tribes.

  • samplingCompleteness.ipynb - accumulation curves of SAGS belonging to different tribes. Used to gauge how completely (or not) we have sampled the pan-genome of different tribes.

  • toDo.ipynb - Wow, I have a lot of stuff to do!

ProcessedModelFiles

Processed versions of KBase reconstructions.

  • AAA023D18.xml - Reconstruction in SBML format
  • AAA023D18RedAdjList.txt - Reconstruction in adjacency list format, used for RE calculations

Also includes results of seed set calculations:

  • AAA023D18SeedSets.txt - list of seed compounds for the genome, one per line. If multiple seed compounds can trade-off, all are listed on a single line.
  • AAA023D18SeedSetsWNames.txt - Same as above, but using metabolite names instead of IDs.
  • AAA023D18SeedWeights.txt - list of all seed compounds with their weights
  • AAA023D18SeedWeightsWNames.txt - same as above, but using metabolite names instead of IDs.

MergedData

Same data as in ProcessedModelFiles folder, but for aggregates of genomes belonging to a single tribe.

About

No description, website, or topics provided.

Resources

Stars

1 star

Watchers

1 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors