The goal of the Center for Computational Regulatory Biology is to develop, refine and test computational approaches in genomics broadly and cis-regulatory analysis specifically. The primary focus for the latter is the elucidation of gene regulatory networks in development. The Center interacts with the wider research community in several ways: it provides open source software for use by academic research groups; it provides web-based servers for genomic analysis using software developed locally; and it maintains databases fundamental to the Sea Urchin Genome Project, an initiative that began in the Davidson laboratory and at the Genomics Technology Facility. The Facility provides to the Caltech and external scientific community upon request services and materials stemming from the macroarray libraries and arraying equipment that we maintain. The Center and the Facility are both under the direction of R. Andrew Cameron. Oversight for the Genomics Technology Facility is shared by the PI of this Center (Eric Davidson) and the PI of the Genomics Research Center (Mel Simon). The work in the facility is conducted by Eve Helguero and Autumn Yuan.

Computational aspects of gene regulatory network project. The description and simulation of gene regulatory networks can only be accomplished with computational tools specific to the task. We have locally developed several software tools that are in constant use by our laboratory investigating sea urchin development, as well as over 110 users working in a variety of other systems (see below). These tools were specifically designed to aid the experimentalist working at the bench and using iterative cycles of experimentation and computation. The software tools include: BioArray, a program that uses macroarray spot data from phosphoimagers to manage intensity and position information; SUGAR, a system to perform, display and correlate large-BAC sequence analyses to aid the experimentalist with the functional analysis of cis-regulatory elements in genomic DNA; SeqComp and FamilyRelations, programs for comparative sequence analysis; and NetBuilder, an environment for creating and analyzing models of gene networks.
In order to make the sequence analysis programs convenient we have installed a web-based facility, the Cartwheel Project, that allows the user to have complete control over the process. Within the circumscribed domain of genomic sequence-based information, Cartwheel provides facilities to organize, analyze, and curate information on the level of individual labs. The analyses are then viewable by programs such as FamilyRelations. The Cartwheel Project is the umbrella term for a bioinformatics infrastructure first developed by C. Titus Brown and now maintained and extended by the computational staff of the Center.As the genome data has expanded, we have found it necessary to install another queuing package, Parasol, to facilitate bulk jobs not covered by Cartwheel.
The equipment that supports the computational efforts of the Center includes two 18 unit Beowulf clusters, a web server for the Sea Urchin Genome Project (SUGP) and several dual processor development machines used by the staff for software construction, testing and maintenance. In the past year we have decommissioned one obsolete Beowulf cluster and installed a new one.
The utility of these comparative sequence analysis facilities is reflected in the user population. At present, the Caltech Cartwheel server, Woodward, has 249 total registered users in 40 lab groups. A total of 22181 jobs have run in the last year for a total of 123 CPU days. The majority of these are Seqcomp and Blast analyses.