Sequana Therapeutics Inc., a fledgling genomics company, saidThursday that it has entered the race to solve one of the thorniestproblems in human genetics _ identifying the gene that causesschizophrenia.

What's more, Sequana thinks it may actually win.

The company has pinned its hopes on the unmatched size of a cohortof families collected by collaborators in New York and Oxford,England, and on the company's automated genotyping apparatus,which can churn out genotypes much faster than those in manyacademic laboratories.

Based in La Jolla, Calif., Sequana's main collaborator, Lynn DeLisi,at the State University of New York at Stoneybrook, has collectedaffected sibling pairs from as many as 300 families, many more thanmost competing research groups, said Jay Lichter, the company'sdirector of genetics.

"I'm very optimistic because Lynn has a large collection of familiesalready collected and phenotyped over 10 years," Lichter toldBioWorld Today. "And our ability to do genotyping here is betterthan any of the other people doing schizophrenia work. We have verysharp statistical geneticists, and capital support from the corporation_ we don't have to stop and write grants, or anything."

The size of the cohort is crucial in the type of study that Sequana, aprivately held company, plans to carry out.

In classic linkage studies, researchers rely on multi-generationalfamilies, in which several members have developed schizophrenia.This study design is convenient because the sample size is relativelysmall, and traits can be tracked through generations. The problem,said Lichter, is that different physicians make the diagnoses, and thatdecades often pass between one diagnosis to another.

"When you have a lot of variation in phenotyping, you addcomplexity you can't control for," Lichter said.

Lichter plans to rely on what are called non-parametric linkagestudies, in which researchers examine affected sibling pairs and theirparents.

For these studies, the genotype of the parents becomes irrelevant.Phenotyping becomes more consistent because siblings are morelikely to be diagnosed by the same doctors and live in the samegeographic area.

The drawback is the large number of families needed to give thestudies adequate statistical power and genetic resolution. DeLisisolved that problem by contributing the fruits of a decade of workwith a cohort of as many as 300 affected sibling pairs.

"Lynn started collecting these families over 10 years ago assumingthe technology would catch up," Lichter said. "She was almostvisionary. Ten years ago, while the idea of using sib pairs wasfeasible the genetic maps weren't ready to allow that kind ofapproach."

The search for the genetic cause of schizophrenia has been litteredwith failures. The most promising scientific report, published adecade ago, linked schizophrenia to a region on chromosome five.The study was announced with great fanfare and accepted uncriticallyby the mainstream press. But the work did not live up to its earlypromise.

Today some 20 leading research groups have embarked on a newround of linkage studies looking for the genetic cause ofschizophrenia. Among them are the Medical College of Virginia inRichmond and Johns Hopkins University in Baltimore.

"Everybody else has [cohorts of about] 100 families," Lichter said."With Lynn, we have 300, which gives us the power to detect linkagewith a high degree of confidence."

He said it will take two to four years for Sequana to identify the gene,if indeed it can. n

-- Steve Sternberg Special To BioWorld Today

(c) 1997 American Health Consultants. All rights reserved.