FIRST COMPLETEMAP OF HUMAN GENOME
By David N. LeffScience Editor
Just in time for the holidays, the Whitehead Institute forBiomedical Research has brought joy to the world ofgenome mappers and disease-gene hunters by generatinga complete map of the entire suite of humanchromosomes.
It heralded this achievement by a press conference inCambridge, Mass., Thursday morning, geared to theformal presentation of its record-breaking genomic mapin today's Science. That paper bears the title: "An STS-based map of the human genome."
STS stands for "sequence-tagged sites," which are shortstretches of DNA marking the location of genes and theirfragments, known and unknown, spread at orderlyintervals along the entire length of the genome. From endto end, the Whitehead map numbers 15,086 DNAmarkers, each defined by a polymerase chain reactionassay.
Its just-completed map spreads these landmarks, 199kilobases apart on average, to cover from 94 to 99 percentof the human genome. It will help the final phase of theU.S. Human Genome Project kick off with the new year.This 15-year program aims to decode the exact sequenceof all 3 billion DNA bases that spell out the geneticowner's manual for building a human being.
Molecular biologist Eric Lander, who heads theWhitehead/Massachusetts Institute of Technology Center,said, "Generating the complete sequence of DNA is themost exciting adventure in modern science. The STSsgive us reference points so we can move through thegenome without losing our way."
The 53 mappers, sequencers, informactics andbioengineers who co-authored the Science paper spenttheir first one and one-half years creating the uniqueautomated hardware and software by which they mappedthe genome in the following one year, ending this pastSeptember.
To pin down the relative positions of their 15,086markers, the collaborators started by constructing threestarter maps, each to a different scale:
* a genetic map, produced at Genethon, a French genomicresearch institution, consists of 5,264 markers used totrace inheritance in human families;
* a first-ever radiation hybrid map, created at theWhitehead, consists of 6,193 landmarks, obtained byscreening large chunks of DNA generated by irradiatinghuman chromosomes;
* the STS-content map of the entire human genome, alsoa first, was produced at Whitehead by screening 10,850markers against human DNA fragments carried in yeastartificial chromosomes (YAC).
The Whitehead team then integrated these three different-scale versions into their final map of the entire genome.
A distinctive departure in their technology was makingextensive use of STSs. These short sequences of DNA,ranging from 200 to 500 base pairs each, needn't bestored in freezers as actual DNA. Rather, they are kept incomputer data bases, as highly accessible virtualfragments of known location, recoverable by PCR assay.
When the team started up their project in 1992, theymanaged 6,000 such PCR assays a day. Since achievingfull automation, they can now turn out 300,000 suchanalyses daily on a robotic machine built in-house, whichthey call `the Genometron'.
Thomas Hudson directs the Whitehead's human genomemapping group. "There are things in our map ofimmediate use to people at large," he told BioWorldToday, "and things useful for the sequencing world."
He explained: "Of immediate use to industry and scienceis that this map makes finding genes much easier.Without a physical map, where we were two years ago,after locating a gene to a certain region of a chromosome,people had to spend years isolating a piece of DNA fromthat region. They had to do its physical map."
Hudson cites as a signal example the gene forHuntington's disease. "It was mapped in 1983, and tookten years to clone. We think," he added, "that mostsimple traits can now be found within a year."
Without waiting to formalize completion of theirmapping project to date, the Whitehead opened access toits data to all comers, through the Internet's World WideWeb.
"Our address on the Web," Hudson said, is(http://www.genome.wi.mit.edu). In one recent week,"Hudson said, "we had 53,000 accesses to that site. Itoffers, free of charge, all the clones, all the markers, so itjust accelerates gene-finding in a major way."
Among his briskest typical demands are requests forgene-hunting material on schizophrenia. "There must besix groups in the U.S. doing that," Hudson observed."Other disease targets have included diabetes andasthma."
"Most of the time," he continued, "people are notinterested in the whole map. They look up some geneticmarker that they've already typed in their families, forinstance, chromosome 6 for schizophrenia; chromosome1 for melanoma."
With its 15,000-plus STS marker map, Whitehead ishalfway to the initial goal set by the U.S. Human GenomeProject, (USHGP) namely, to establish a physical map ofthe human genome with 30,000 STS landmarks by 1998.Thereafter, the second phase will move from mapping tosequencing.
"Everyone in the U.S.," Hudson said, "is poisingthemselves, getting ready to go for the sequencingproject. It has two challenges. One, you can have a highthroughput sequencing operation, but, two, you also haveto be able to select the right clones to sequence."
The Whitehead team, Hudson suggested, can reach theUSHGP's 1998 goal of 30,000 markers by the end of1996. "That doesn't seem a big challenge any more," hesaid. "Our next objective is to become one of the toplarge-scale sequencers; to sequence a large percentage ofthe human genome."
That ambition is grounded in Whitehead's hyper-automated technology. Most of it," Hudson observed,"has been in developing robots to prepare the DNA, pickthe clones, purify the DNA and do the sequencingreactions." He added, "The robotic systems we havedeveloped here are unique. No one else has them."
A healthy, anonymous adult male, Hudson related, atFrance's Centre d'Etude du Polymorphisme Humain(Human Polymorphism Research Center) provided theblood cells for the prototype human clone library thatlaunched the international Human Genome MappingProject in 1992. n
(c) 1997 American Health Consultants. All rights reserved.