FileDetails.txt

The following is a brief summary of the files available in this repository and are related to the manuscript, "Gene networks provide a high-resolution view of bacteriophage ecology" by Jason W Shapiro and Catherine Putonti.

Additional information on how to use the data and code provided is described in the file "Walkthrough.txt."

Directories and files in "Output Data":
centroids35DB: contains a protein blast database of all centroid sequences of the phage FAA file clustered with usearch at 35% identity
edgelists: contains all edgelists in two-column format. All files are text files; files with the extension '.abc' were used as input for MCL-edge
	-files with the prefix 'gedge35' pertain to gene-level networks
	-files with the prefix 'pedge35' pertain to genome-level networks
	-the suffix 'addedgeN' means that N random edges were added per node to generate the given edgelist
	-the suffix 'droptop' means that the top 3% connected nodes (by degree) were dropped to test the effect on clustering
	-the suffix 'unique' means that only one direct of each edge ('node A' to 'node B' is written to the file)
MCL_genes: contains the result file ('.dump') from running MCL on the associated gene network
	-the value 'I##' refers to the inflation factor (multiplied by 10) used to generate each MCL clustering. The extension .dump is used as reference to the step in MCL-edge
	-other prefixes and suffixes as described for 'edgelists'
MCL_genomes: contains the result file ('.dump') from running MCL on the associated genome network
	-prefixes and suffixes as defined in 'edgelists' and 'MCL_genes'
mimax_output: result files from running the mimax algorithm 10 times with the set of clusters 'gedge35_addedge5.I150.dump' (in MCL_genes folder)
	-the prefix 'mimat' defines files with the final cluster-host matrix after mimax
	-the prefix 'mipres' provides the final genome-gene presence/absence matrix after mimax
	-the prefix 'mival' provides the vector of mutual information scores after each iteration of mimax
	-the numerical suffix refers to the replicate of running mimax on the same set of MCL clusters
genustree.tree: Newick format of genus-level tree of phage hosts

Files in "Input Data":
aligncat.fa: Concatenated alignments of single-copy marker genes across representative host genomes associated with phages in the dataset.
allphage.faa: FAA file with all phage genomes concatenated with headers in the format "Accession_N" where "Accession" refers to the phage genome accession in GenBank, and "N" refers to the Nth gene in that genome.
hostspectree.tree: species-level tree of representative hosts, labeled by host GenBank accession
markerDB.tar.gz: compressed directory of single-copy bacterial gene BLAST database
MuscleRelabel.tar.gz: compressed directory of MUSCLE alignment results, relabeled with just host genome accessions. Files named for the bacterial genes.	
phagehostdatasummary.txt: tab-delimited table with headers 'Phage Accession', 'Host Species', 'Host Genus', 'Used in analysis' (1 = yes, 0 = no), 'Phage Family', and 'Dataset' (Network = part of original network; Random Sample = part of random test)
phagehosts_genus.txt: tab-delimited table with phage genome accession in the first column, and host genus in 2nd. No headers.
phagehosts_species.txt: as phagehosts_genus.txt but the 2nd column contains hosts defined at the species level.
phagehosttable.txt: supplementary table (not used in Walkthrough) with phage accession, host strain name, and representative host accession
phagelist.txt: list of phage accessions (includes all 945 in the allphage.faa file, not just those remaining after usearch)
	
Files in "R Scripts":
alignpaste.r: script for concatenating alignments
CIplot.r: R script for generating a regression plot with confidence intervals
GeneNetworkRscripts.r: set of R scripts used in the study. Usage is explained in "NetworkMethodsSummary.txt" and briefly described in the file itself.
hostassign.r: set of functions for predicting phage host information and for creating a vector of host associations for each phage gene or genome