Biorithm: Process GO

Introduction

The process-go tool can be found in the process-go directory of Biorithm. process-go is built around modest needs to parse and manipulate Gene Ontology (.obo) files. process-go currently recognizes all tags mentioned in the GO format description. Any new tags will generate warnings echoed to standard error and will be ignored. Certain tags are currently not found in GO files. Some of these generate warnings as well.

process-go currently supports fast, efficient access to data in the obo files. The current obo file takes about 5 seconds to preprocess with process-go, and after that all data requests are very fast.

Installing Process GO

Download the Biorithm package and follow the installation instructions for Biorithm. The executable files will be available as process-go/process-go

Command Line Options

The following options are availble for process-go:

Flag	Long	Description
a	annotations-file	The name of the file containing functional annotations. The first line of the file specifies the column name. Each tab-delimited line contains the gene id (in column 'orf'), the GO id (without the GO:0+ prefix, in column 'goid'), the GO category (in column 'hierarchy'), the evidence code (in column 'evidencecode'), and the annotation status (1, 0, or -1, in column 'annotation type').
A	ancestors	Compute the ancestors of a given set of functions. Use the --functions-file option to specify these functions.
D	depths	Print the depth of each function in GO. The name of the output file is the go-file argument concatenated with the string '-depths'.
S	descendants	Compute the descendants of a given set of functions. Use the --functions-file option to specify these functions.
d	direction-transitive-closure	Specify the direction in which to transmit functional annotations to perform the transitive closure. Use this option only with the --annotations-file and the --transitive-closure options. There are four possibilities: (i) up: transfer annotations up the GO DAG. (ii) down: transfer the annotation status of 'unknown' (0) down the GO DAG, from the most specific annotated function for a gene to all the descendants of the function. (iii) side: in addition to 'down', also transfer annotation status of 'unknown' to all siblings of the function, where two functions are siblings if they share both a common ancestor and a common descendant (iv) updown: both 'up' and 'down' (v)all: 'up', 'down', and 'side'
f	functions-file	The name of the file containing the IDs of functions to process, one function ID per line.
g	go-file	The name of the file containing the structure of the Gene Ontology in OBO format.
o	output-file	The name of the file to print annotations to. You must provide this option with the --annotations-file and --transitive-closure option. In this case, process-go prints the transitive closure of the annotations to the output file.
t	transitive-closure	See the documentation for --output-file option to see what process-go does when also given the --transitive-closure option. Without the --output-file option, compute the transitive closure of the GO DAG and print the relations in the closure. The name of the output file is the go-file argument concatenated with the string '-relations-closed. Each line of the file contains the id of a function, the id of a descendant, and the length of the shortest path between the two functions in the GO DAG.
r	relationship-file	The name of the file to print relationships to. You must provide this option with the --go-file. In this case, process-go prints the relationships of annotations to the output file as a three column tab delimited text file. The file has structure as FunctionA FunctionB Relationship.