Review banner (jpeg, 22K)

bulletORNL Review Home Page
bulletFeatured in This Edition
bulletLast Article
bulletNext Article
bulletSearch the ORNL Review Site
bulletComment on this article

ORNL supports DOE's human chromosome sequencing efforts by computationally analyzing DNA sequences to locate known genes, predict unknown genes, and estimate gene functions.

What's in a Chromosome?
Tune in to the Genome Channel

In April 2000, in Walnut Creek, California, researchers from three DOE national laboratories completed the draft sequencing of human chromosomes 5, 16, and 19 for the Human Genome Project. These were not the first chromosomes to be sequenced; the first was chromosome 22, whose order of DNA bases was published in December 1999 by a group of institutions funded by the National Institutes of Health (NIH). However, these were the first human chromosomes to be sequenced by DOE's Joint Genome Institute (JGI), which integrates the genome centers at the Lawrence Berkeley, Lawrence Livermore, and Los Alamos national laboratories with informatics and functional genomics capabilities at ORNL.

ORNL supports JGI by computationally analyzing DNA sequences to locate known genes, predict unknown genes, and estimate the functions of many of these genes. It is known that chromosomes 5, 16, and 19, which make up roughly 11% of the human genome, contain genes that are linked to kidney disease, various cancers, hypertension, and diabetes.

In addition, through its Web site, called the Genome Channel, ORNL provides new information every month on the latest DNA sequences obtained experimentally (including partial and whole sequences of human chromosomes) and their computationally identified genes.

The Genome Channel is a Web browser tool that provides a comprehensive sequence-based view of parts of the human and mouse and several complete microbial genomes. This information resource is being used by many people worldwide, ranging from biomedical researchers at NIH and the Massachusetts Institute of Technology to scientists who are developing drugs at some of the world’s largest pharmaceutical companies. The Genome Channel, which has about 100,000 hits per month, was listed among five "hot pick" Web sites in the December 17, 1999, issue of Science magazine.

A view of the Genome Channel home page on the World Wide Web.

A researcher interested in developing a drug to control Parkinson's disease or cancer can get relevant information from the Genome Channel by typing in a class of genes, such as "caspase," in the search space. "Caspase genes control cell death," says Ed Uberbacher, head of the Computational Biosciences Section in ORNL's Life Sciences Division. "In Parkinson’s disease, cells die when they shouldn't, and in cancer, cells grow and spread instead of dying when they should."

Regulatory proteins are of interest to researchers because they control the activity of a host of genes, turning some on and others off, thus affecting the types and levels of proteins produced. The Genome Channel provides information on regulatory proteins as well as up-to-date lists of many important gene types, based on the most recent ORNL analysis.

Uberbacher was one of the developers of the Gene Recognition and Analysis Internet Link (GRAIL) computer program. Among other things, GRAIL is known for having helped researchers locate the gene responsible for Adrenoleukodystrophy (related to the disease featured in the movie Lorenzo's Oil). The updated version of GRAIL, called GRAIL EXP, is available to users of the Genome Channel.

"Using ORNL's IBM SP supercomputer, our programs search the databases for expressed sequence tags," Uberbacher says. "These are fragments of gene sequences that resemble or match DNA sequences we are interested in analyzing. Using these sequences as evidence, we can identify and locate genes along the human genome. By knowing the gene sequence, we can then predict the amino-acid sequence of the gene's protein product. By matching the sequence to known protein sequences in a database, we can often predict the biochemical function of the protein.

"We link gene sequences computationally to genetic maps of chromosomes showing markers, or known locations along a chromosome, that are tied to known biological functions. By doing this data mining, we can predict a function for a particular gene sequence of interest."

According to Uberbacher, ORNL researchers will estimate two-thirds of the functions of the genes that make up the sequenced chromosomes. "We will be able to predict that a gene's protein product catalyzes a particular biochemical reaction," he explains. "But to find out what the gene actually does in the organism, we will rely on ORNL biologists studying the effects of mutations of a similar gene in the equivalent chromosome in mice. For example, the biologists might find that the gene affects embryonic development."

To get updates on the latest findings concerning human chromosomes 5, 16, and 19, stay tuned to the Genome Channel (compbio.ornl.gov).

Beginning of Article

Microbial Functional Genomics and Waste Site Bioremediation Table of Contents Search the ORNL Review Site Comments to Editor ORNL Review Home Page ORNL Home Page