The Genome Analysis Unit

he recent, explosive growth of genomic and proteomic data has dramatically changed the face of biomedical research. The formal scientific discipline of bioinformatics has emerged to address the formidable challenges associated with storing, analyzing, and integrating genetic and other biological information through computer technology. While opening many doors for new areas of investigation, bioinformatics and its associated deluge of data also present many new challenges—challenges that bench researchers are often ill equipped to face. Many bioinformatics approaches are now necessary components of modern molecular biology research, and just as changes in sequencing technology mobilized the field of molecular biology to move more away from sequencing a single gene to sequencing an entire genome, so bioinformatics is now more frequently being used to address issues involving large classes or families of genes rather than single genes or proteins.

At the CCR, the Genome Analysis Unit (GAU) was created to address some of the issues resulting from the explosive growth in genomic data and to provide a central resource to enhance the research productivity of CCR scientists. The GAU serves this function through a variety of avenues, such as collaborative projects, developing general-purpose web tools, and presenting and organizing training seminars. This article focuses on two of its projects: 1) the NCI Bioinformatics Community Resource, which is a rating guide for using particular web-based tools, and 2) a collaborative project in which bioinformatics was used to help bench researchers locate a small non-coding regulatory RNA.

The NCI Bioinformatics Community Resource

While the Internet has provided researchers with unprecedented access to repositories of data, literature, analysis tools, and other assorted research information, making effective use of these tools is far from straightforward. In fact, “information overload” is a major problem. Simply keeping track of all the resources can be a full-time job. For example, “The Molecular Biology Database Collection” (Nucleic Acids Res, 2005, vol. 33 [Database issue]: D5–D24) listed no fewer than 719 different databases ranging from the well-known databases, such as GenBank (an annotated collection of all publicly available nucleotide and protein sequences), to lesser-known systems such as the Aptamer database (a collection of small RNA or DNA molecules capable of binding ligands, ranging from small organic compounds to whole organisms)! Not only is the number of databases overwhelming, but making efficient and effective use of them is made more difficult by the fact that different viewing and analysis tools may exist at different sites. To address this issue, the GAU has recently launched the NCI Bioinformatics Community Resource (NBCR) (http://genome.nci.nih.gov/nbcr). The NBCR is a repository (database) of links to an array of bioinformatics resources useful in the analysis of DNA and protein sequence data. Designed to be a community-managed resource, researchers are invited to provide meaningful reviews of the listed sites and suggestions for new sites. The goal is to construct a rating guide via peer review of those resources that may prove valuable to the NCI community and to provide direction on how best to navigate the ever-growing sea of information associated with those resources. We expect that the NBCR will be a uniquely valuable tool via a rating scheme that reflects “real-world” utility.

Collaborative Research and Custom Tool Development

During the past year, the GAU has been involved in a number of successful collaborative projects with several CCR scientists. The focus of these projects has included custom oligo design for micoarray chip production, development of simple web-based tools for identification and extraction of promoter regions, development of gene annotation and DNA codon modification tools, a genome-wide analysis of human promoter regions, and the search for small regulatory RNA (sRNA) candidates in both eukaryotic and prokaryotic organisms. These collaborations have been quite successful in moving NCI science ahead. One such collaborative project is detailed below.

Identification of Tandem Duplicate
Regulatory Small RNAs in Pseudomonas aeruginosa Involved in Iron Homeostasis

Wilderman PJ*, Sowa NA^‡, FitzGerald DJ^‡, FitzGerald PC^‡, Gottesman S^‡, Ochsner UA*, and Vasil ML*. Proc Natl Acad Sci U S A 101: 9792–7, 2004.

* University of Colorado Health Sciences Center, Denver, CO
^‡ National Cancer Institute, National Institutes of Health, Bethesda, MD

Small non-coding RNAs (sRNA) are located predominantly in the intergenic (IG) regions of bacterial genomes. One of the challenges in understanding their contribution to gene regulation has been simply locating them. Previously, Massé and Gottesman 2002 (Proc Natl Acad Sci U S A 99: 4620–5) demonstrated that the expression of a Fur-regulated sRNA (RyhB) is responsible for the regulation of an assortment of genes in Escherichia coli that are expressed under iron-replete conditions. Sequence homologs of these sRNAs were also identified in other Enterobacteriaceae (e.g., Salmonella, Klebsiella, and Shigella). However, this sequence homology did not extend to the genus Pseudomonas. Because the vast majority of the sRNAs that have been described in E. coli are encoded in IG regions and no microarray chips are available that cover the IGs of Pseudomonas, a different approach was needed. Thus, a RyhB functional homolog was sought by querying all the IG regions, derived from the whole genome sequence (GenBank id: NC_002516), of the PAO1 strain of P. aeruginosa for two predicted properties of such a functional homolog: regulation by a Fur box, and a rho-independent transcription terminator. This analysis yielded only three candidates. Two of the candidates (now termed PrrF1 and PrrF2) were located in tandem between the genes PA4704 and phuW (PA4705). Microarray and expression studies, as well as gene deletion experiments, demonstrated that both members of this tandem pair are Fur- and iron-regulated, and that they are functional, but not sequence, homologs of RyhB. Moreover, while homology searches found two putative prrF sequence homologs in P. putida, P. fluorescens, and P. syringae, they are considerably distal to each other in these organisms in contrast to their tandem location in P. aeruginosa. We conclude from this study that this type of bioinformatics approach is likely to be successful in finding other sRNAs regulated by any well-defined regulatory protein in any sequenced organism that is known to use rho-independent terminators.

In conclusion, the above-mentioned project is just one example where, by bridging the gap between molecular biology and bioinformatics, the GAU has collaborated with CCR scientists to produce a successful outcome not as easily achieved by either partner alone.

NCI scientists wishing to contact the GAU can do so by sending an email to pcf@helix.nih.gov or by visiting our web site at http://genome.nci.nih.gov.

Peter FitzGerald, PhD
Head, Genome Analysis Unit
Office of Science and Technology Partnerships
Office of the Director
NCI–Bethesda, Bldg. 37/Rm. 2012
Tel: 301-402-3044
Fax: 301-402-3044
pcf@helix.nih.gov

back to top