HomeResearchIntramural ResearchResearch Branches at NHGRIGenome Technology Branch Mullikin Lab

Jim Mullikin, Ph.D. Associate Investigator Genome Technology Branch Head Comparative Genomics Unit B.S. Purdue University, 1982 M.S. Purdue University, 1984 Ph.D. Delft University of Technology, Holland, 1993 (301) 496-2416 (301) 480-0634 mullikin@mail.nih.gov 5625 Fishers Ln Room 5N-01Q, MSC 9400 Rockville, MD 20892-9400 Selected Publications

Dr. Mullikin develops and utilizes computer programs to analyze large data sets generated through systematic DNA sequencing projects. A highly skilled computational geneticist, he collaborates extensively with other genomic researchers, analyzing data others collect or that are available in public databases.

His main work involves creating algorithms for performing complex computations. He designed one such program, called Sequence Search and Alignment by Hashing Algorithm (SSAHA), to dramatically accelerate the speed at which gigabases of DNA sequence are searched for single nucleotide poymorphisms (SNPs). It was developed several years ago, and Dr. Mullikin is still refining SSAHA in response to the continually changing needs of genomic scientists. SSAHA remains the key tool that he and others use to detect genetic variation. He also developed a program called Phusion—pronounced "FUSION"— for assembling a genomic sequence from a whole-genome shotgun data. The genomic sequences of both mouse and the nematode (Caenorhabditis elegans) were assembled using Phusion.

Dr. Mullikin’s group also provides computational support to major NHGRI efforts, such as the International Haplotype Map (HapMap) Project, which is primarily focused on determining genes and genetic variations that affect health and disease susceptibility. In the initial phase of the project, investigators aim to produce a working HapMap consisting of about 600,000 polymorphic sites roughly five kilobases apart from one another. Some particularly complex regions will need even higher resolution. When completed, the HapMap Project will give researchers a powerful tool for isolating and identifying disease genes through association studies.

In addition, Dr. Mullikin is developing algorithms for NHGRI’s comparative sequencing efforts, which involve the sequencing of many additional vertebrate genomes to help annotate the human sequence. Because only one percent to two percent of the human genome codes for protein, comparing the human genome with that of other vertebrates is key to highlighting other regions that are functionally important. In humans and mice, for example, as much as three percent of the noncoding sequence seems to have been conserved. Dr. Mullikin and other scientists are working to determine what these conserved, noncoding sequences are doing; some probably are involved in gene regulation, while others may have structural roles.

In one project, he is developing signal processing algorithms for sorting out the confusion that can result when polymerase chain reaction technology is used to amplify a segment of the genome for subsequent SNP discovery. During such a process, two copies of every chromosomal region (except for those on the X and Y chromosomes in males) can get sequenced. This can be confusing in heterozygous cases, where the regions contain different SNPs. Simply allowing for the possibility of a heterozygous state is not the answer; it merely increases the error rate. Dr. Mullikin is trying to devise software for unraveling the signals so the correct information can be captured. To take the algorithm a step farther, he is working on methods for reliably detecting insertion-deletion polymorphisms—situations where an entire sequence is inserted or deleted, which creates a much more complicated signal than does an individual base change. His work should make it possible to extract that information from the mixed signals.

Top of page

Last Updated: August 1, 2008