Hubber

Hubber calculates the cohesiveness of one or more input gene sets in a given functional relationship (or experimental data) network; it also calculates the association of each gene with those gene sets. This is primarily useful for predicting function for genes based on "guilt by association" with genes already known to be in those functions (i.e. gene sets).

Usage

Basic Usage

 Hubber -i <network.dab> -g <genes> <contexts.txt>*

Output (to standard output) the average in- and out-connectivity of each gene set contexts.txt in the network network.dab; for each context, display the genes genes most specifically associated with that context. In-connectivity is the weight of an edge between two genes in the context; out-connectivity is the weight of an edge incident to a gene in the context.

 Hubber -i <network.dab> -g -1 <contexts.txt>*

Output (to standard output) the specificity with which every gene in the genome is associated with each context contexts.txt in the network network.dab. Scores for genes already annotated in the context are negative.

 Hubber -i <global.dab> -o <backgrounds.bin> -x <context_ids.txt> -s <gene_ids.txt>
        -d <contexts_dir> <contexts.txt>*

Save in backgrounds.bin the average weight of edges incident to each gene in the global network global.dab and in every context-specific network in contexts_dir, each corresponding to a context gene list contexts.txt. context_ids.txt and gene_ids.txt are tab-delimited text files as described below (and used in BNServer) with one-based indices in the first column and context or gene names, respectively, in the second column.

Detailed Usage

package "Hubber"
version "1.0"
purpose "Inspect a DAT/DAB file for gene network properties"

section "Main"
option  "input"     i   "Input DAT/DAB file"
                        string  typestr="filename"
option  "genes"     g   "Number of genes to output per gene set"
                        int default="0"
option  "weights"   w   "Input weights PCL file"
                        string  typestr="filename"

section "Miscellaneous"
option  "normalize" n   "Normalize input file"
                        flag    off
option  "genin"     e   "Gene inlusion list"
                        string  typestr="file"
option  "genex"     G   "Gene exclusion list"
                        string  typestr="file"
option  "clip"      c   "Output directory to clip subnetworks for each gene set"
                        string  typestr="directory"

section "Backgrounds"
option  "output"    o   "Output gene backgrounds"
                        string  typestr="filename"
option  "contexts"  x   "Contexts for gene backgrounds"
                        string  typestr="filename"
option  "genelist"  s   "Genes for backgrounds/withins"
                        string  typestr="filename"
option  "directory" d   "Context network directory"
                        string  typestr="directory" default="."
option  "genesets"  t   "Gene sets for within calculations"
                        string  typestr="filename"
option  "between"   b   "Gene sets for between calculations"
                        string  typestr="filename"

section "Optional"
option  "threads"   a   "Number of simultaneous threads to use"
                        int default="1"
option  "memmap"    m   "Memory map input"
                        flag    off
option  "verbosity" v   "Message verbosity"
                        int default="5"

Flag	Default	Type	Description
None	None	Gene text files	Gene sets which will be tested for cohesiveness and gene associations in the given dataset.
-i	stdin	DAT/DAB file	Interaction network which will be analyzed for gene/function association.
-g	0	Integer	Number of genes to output in association with each gene set. 0 will display no genes, only gene set cohesiveness scores, and -1 will produce a table of every gene's association with every function.
-n	off	Flag	If on, normalize input edges to the range [0,1] before processing.
-c	None	Directory	If given, for each gene set, produce a DAB file in the given directory containing only the genes (and edges) in that gene set.
-o	None	Binary file	If given, binary file into which each gene's average connection weight to every other gene in each context is placed. Used with BNServer.
-x	None	Text file	If -o is given, tab-delimited text file containing at least two columns, the second of which must be a context name corresponding to that context's functional relationship DAT/DAB file. Identical to the BNServer context list.
-s	None	Text file	If -o is given, tab-delimited text file containing at least two columns, the second of which must be a gene name; should list all genes for which backgrounds are to be calculated. Identical to -x in format.
-d	.	Directory	If -o is given, directory from which DAT/DAB files with names corresponding to contexts from -x are loaded.
-m	off	Flag	If given, memory map the input files when possible. DAT and PCL inputs cannot be memmapped.