Dr. Collins' research laboratory is focused on the identification and understanding of genes that cause human disease, using methods of positional cloning. Past efforts have been successful in identifying the genes for cystic fibrosis, neurofibromatosis, and Huntington disease. An intense NIH collaboration over the past three years, led by Dr. Collins, succeeded in identifying the gene for multiple endocrine neoplasia type I (MEN1) in 1997. The gene, which meets all formal criteria for a tumor suppressor, has no homology to any other previously identified DNA sequence. Antibody tagging experiments have recently demonstrated that it is a nuclear protein, and efforts are underway to define its function.

The Collins lab is also attempting to extend this genetic analysis approach to more difficult non-Mendelian problems. A major effort is underway to pursue the genetic basis of non-insulin dependent diabetes mellitus (NIDDM), by studying a large cohort of affected sib pairs and relatives collected in Finland. This project involves a whole genome search for genes conferring susceptibility to diabetes, or intermediate traits such as insulin resistance, in 2,500 DNA samples. A total of 1,000,000 genotypes have been performed in phase I of the study, leading to the conclusion that a major gene for NIDDM lies on Chromosome 20. Several other potential gene locations are being followed up by a replication study and linkage disequilibrium analysis in this unique population.

Many of the genes conferring risks for disease are large, and a wide variety of mutations are capable of conferring phenotypic consequences. Thus for successful introduction of genetic analysis into clinical medicine, the technology for mutation detection must be capable of detecting all possible mutations in very large targets, at high sensitivity and specificity, and at affordable cost. In collaboration with Affymetrix, we are developing methods to use DNA chips for detection of heterozygous mutations in the BRCA1 gene, a good example of a large gene with almost limitless mutational possibilities. Using two-color fluorescent methods, sensitivity and specificity of >90% have been achieved. This technology is now being extended to other genes such as BRCA2 and the ataxia telangiectasia gene, and to an evolutionary analysis in higher primates.