GoMiner Command Line Syntax

Application Build: 246 Database Build: 2008-04

Home

GoMiner™ Command Line Syntax

System Requirements

Computer with Java 1.4 or higher installed
The application has been tested on Windows 98, NT, 2000, XP, Solaris, Linux, FreeBSD and Mac OS X
The application does not run on Mac OS 9, since there is no java 1.4 support for this platform
High speed internet connection (if your connection is too slow, you can configure GoMiner to use a local database instead)

Command Line Syntax

Command Line:

java -cp gominer.jar [Optional JVM parameters] gov.nih.nci.lmp.gominer.GOCommand [GoMiner parameters (see below)]
Optional JVM Parameters (Use if Gene Files Exceed 10,000 Genes or if you get an OutOfMemory exception): -Xms256M -Xmx256M
Required GoMiner Parameters:

Parameter Flag Expected Values Example (Windows Format)

Total Gene File -t Absolute or relative path to file containing the list of all of the genes, typically all of the genes on your microarray. If you have no total file, you can type GENERATE for the system to generate a total file (please be sure to specify an organism and preferably a data source) -t C:\MyFiles\AllMyGenes.txt or GENERATE

Changed-genes file -c (or) -h -c is used to specify a single changed-gene file. Changed-gene file should contain lists of all of the changed genes to be analyzed, typically from a microarray experiment
-h is used to specify an input file containing list of changed-gene files -c C:\MyFiles\ChangedGenesExp1.txt or -h C:\MyFiles\ChangedGenesFileList.txt

Note:"ChangedGenesFileList.txt" contains list of changed-gene files.

C:\MyFiles\ChangedGenesExp1.txt C:\MyFiles\ChangedGenesExp2.txt C:\MyFiles\ChangedGenesExp3.txt

Database -d JDBC URL. This information is either the connection information for the discover server, or from your own database, if you have configured a local database -d jdbc:mysql://discover.nci.nih.gov:1521/GEEVS

JDBC Driver

-j

Java class name for the JDBC driver class for the database you are accessing. This information is either the connection information for the discover server, or from your own database, if you have configured a local database. The two supported drivers are:

com.mysql.jdbc.Driver
org.apache.derby.jdbc.EmbeddedDriver

-j com.mysql.jdbc.Driver

Database Username -u The database user name. This information is either the connection information for the discover server, or from your own database, if you have configured a local database. If you are using Derby as the local database, use the value NONE. -u deploy

Database Password -p The database password for the specified user name. This information is either the connection information for the discover server, or from your own database, if you have configured a local database. If you are using Derby as the local database, use the value NONE. -p selectonly

Results Directory -r A relative or absolute path to a directory into which the results will be placed. If the directory does not already exist, one will be created. -r C:\MyFiles\MyResults

Data Source

-s

Which data source(s) should be used for making GO associations. The available options are either all or a semi-colon delimited list of specific data source(s).
NOTE: On some Unix shells, it is necessary to escape the semi-colon with a backslash, i.e. use \; instead of ; as the delimiter.
The available choices are:

CGEN (H. sapiens et al.)
SPTR (H. sapiens et al.)
UniProtKB (H. sapiens et al.)
TIGR_TGI (H. sapiens et al.)
TIGR_CMR (Microbes)
FB (D. melanogaster)
GeneDB (G. morsitans)
GeneDB_Tbrucei (T. brucei)
GR (GR)
MGI (M. musculus)
RGD (R. norvegicus)
SGD (S. cerevisiae)
TAIR (A. thaliana)
TIGR_Ath1 (A. thaliana)
TIGR_Tba1 (T. brucei)
WB (C. elegans)
ZFIN (D. rerio)

-s all or -s MGI or -s MGI;RGD

Organism -o Which organism(s) data should be used for making GO associations. The available options are either all or a semi-colon delimited list of NCBI taxonomy id's.
NOTE: On some Unix shells, it is necessary to escape the semi-colon with a backslash, i.e. use \; instead of ; as the delimiter. -o all or -o 602 or -o 602;10090

Export Type -e Which export file format(s) should be generated. The user many specify one or more export format types in a semi-colon separated list. The available options are [se][gce][svge][vsvge][fdrse]. The se option creates a summary for each category. The gce option lists all of the genes in each category. The svge option creates a DAG (Directed Acyclic Graph). The vsvge option creates a VennMaster view of changed category. The fdrse option creates a summary for each category including FDR values. -e se

Root Category

-a

Which category should be used as root for the export. The available options are either all or GO ID. The popular root categories include:

biological_process: GO:0008150
cellular_component: GO:0005575
molecular_function: GO:0003674

-a all or -a GO:0008150 or -a GO:0005575
 or -a GO:0003674

Minimum Category Size for Selecting Randomized Categories -m Limits the categories that will be included in the CIM and summary reports. Categories whose size is less than this threshold will be omitted from category statistic calculations, and randomized categories below this threshold will be omitted from FDR calculations. -m 5 or -m 1

Evidence Code

-v

Which sets of Evidence Codes to use as the criteria for accepting gene to GO category mappings. Based upon our own experience, we recommend level 3 for general use.

Level

all

-v all or -v LEVEL3 or -v TAS;IDA

Use Enhanced Name Column -n Specifies how user submitted gene identifiers should be looked up from the database.
The available choices are

[true] : Use our enhanced H. sapiens et al. gene name association . This option only has a functional meaning if H. sapiens et al. is included for the –Data Source option.
[false] : Do not used the enhanced H. sapiens et al. gene name column -n true

Use Cross Reference Table -x Specifies how user submitted gene identifiers should be looked up from the database.
The available choices are

[true] : Includes the GO Consortium data from GO's cross reference table

[false] : Do not use the cross reference table -x true

Use Synonym Table -y Specifies how user submitted gene identifiers should be looked up from the database.
The available choices are

[true] : Includes the GO Consortium data from GO's gene product synonym table

[false] : Do not use the gene product synonym table -y true

Number of Randoms -f Number of randoms to be used in FDR calculation.
This option is required only if Export Type is fdrse -f 25

Complete Example for Windows:

java -cp gominer.jar gov.nih.nci.lmp.gominer.GOCommand -t C:\gominer\total.gene.txt -c C:\gominer\changed1 -d jdbc:mysql://discover.nci.nih.gov:1521/GEEVS -j com.mysql.jdbc.Driver -u deploy -p selectonly -r C:\dumptest -s UniProtKB;MGI -o all -e gce;svge -a all -m 5 -v all -n true -x false -y false -f 25

Complete Example for Unix:

java -cp gominer.jar gov.nih.nci.lmp.gominer.GOCommand -t /usr/home/gominer/total.gene.txt -c /usr/home/gominer/changed1 -d jdbc:mysql://discover.nci.nih.gov:1521/GEEVS -j com.mysql.jdbc.Driver -u deploy -p selectonly -r /usr/home/gominer/dumptest -s UniProtKB\;MGI -o all -e gce\;svge -a all -m 5 -v all -n true -x false -y false -f 25

Notes

Command line flags and parameters are space-delimited
The list of parameters associated with the -c flag can be of arbitrary length
The parameters associated with the -s and the -o flags are separated by a semicolon and no spaces
The output files are located in RESULT_DIRECTORY
The output file names consist of the original CHANGED_FILE name with an automatically-appended suffix consisting of the date and the EXPORT_TYPE
We have deployed an enhancement for the gene name annotation for the UniProt data in the GO Consortium database. The traditional GO Consortium database is accessed by using the -n false option; the enhanced human annotation is accessed by using the -s UniProtKB option AND the -n true option
When the parameter associated with the -s flag is UniProtKB, the input genes are interpreted as gene names when available and the associations are made using the column added as part of the enhancement.

Limitations

Although an arbitrary number of CHANGED_FILE names is permitted, only a single TOTAL_FILE is permitted. This TOTAL_FILE must in fact be the correct (i.e. a superset) TOTAL_FILE for each of the CHANGED_FILE names
Only one type of export file can be specified; the command line needs to be run once for each type of expport, i.e. once with the se option, once with the gcem option, and once with the svge in order to obtain all of the types of exports.
The export files contain all the categories of the Gene_Ontology; unlike the GUI version of GoMiner a sub-portion of the entire Ontology cannot be selected
If the gce option is specified, the command line will need to include the TOTAL_FILE in the semicolon-delimited list of CHANGED_FILEs if the user wishes to compare the categorization of the CHANGED_FILE genes and the TOTAL_FILE genes.

Using the GoMiner Command Line Interface for High-Throughput Processing

The command line interface can be used to include other components as part of high-throughput processing. A typical processing scenario could include the following steps in an automated script:

A statistical operation identifies a set of changed genes
The set of genes if formatted into the GoMiner changed-gene file format
GoMiner processes the files using the command line interface
The results exported from GoMiner are analyzed
The user is notified about which changed-gene files are of interest

After you have identified results of particular interest, you can use the GoMiner GUI tool to examine them.

We would like to hear from you. You can reach the team via email.

GoMiner was originally developed jointly by the Genomics and Bioinformatics Group (GBG) of LMP, NCI, NIH and the Medical Informatics and Bioimaging group of BME, Georgia Tech/Emory University. It is now maintained and under continuing development by GBG.

Notice and Disclaimer