Andrew Su, Ph.D.
Senior Research Investigator I
In the Computational Biology Group, we are particularly interested in the development and application of bioinformatics algorithms for research in functional genomics and genetics. As experimental science at GNF involves continual innovation, our group’s analysis and modeling of data are also constantly evolving. Generally speaking, however, our research falls in two main areas: first, gene function and regulation, and second, genetics and disease.
Gene Function and Regulation
We have a long-standing interest in using modern functional genomics technologies to understand gene function and gene regulation. Most notably, we have used high throughput gene expression analysis to generate a gene atlas of mouse and human expression patterns across diverse tissue sets (available at http://symatlas.gnf.org). This atlas provides an informative level of genome-wide cellular gene annotation to complement descriptions of molecular gene function derived from sequence analysis. We have also used high throughput gene expression analysis in other types of studies, ranging from hypothesis generation (implicating the role of lipid metabolism in HCV) to pattern recognition (cancer classification, circadian rhythms). More recently, we’ve developed algorithms to analyze data from high throughput library screening of siRNA and cDNA collections, and we are investigating algorithms to infer gene networks from all of these high throughput data sources.
Genetics and Disease
Leveraging the highly parallel SNP-discovery and genotyping capabilities at GNF, we have developed algorithms to perform haplotype association studies using a diverse panel of inbred mouse strains. This approach has been validated in several test cases where the genes or loci that influence disease or phenotype are known. We are currently exploring this algorithm and mouse population in the “genetical genomics” or eQTL context, in which gene expression measurements are used as phenotypes. We expect that the combination of genotype, phenotype, and gene expression (and in the future, other genome-wide profiling) techniques will provide a more comprehensive picture of disease and gene networks.
Selected Publications
- McClurg P, Janes J, Wu C, Delano DL, Walker JR, Batalov S, Takahashi JS, Shimomura K, Kohsaka A, Bass J, et al. Genomewide association analysis in diverse inbred mice: power and population structure. Genetics 2007;176(1):675-83.
- Su AI, Hogenesch JB. Power-law-like distributions in biomedical publications and research funding. Genome Biol 2007;8(4):404.
- McClurg P, Pletcher MT, Wiltshire T, Su AI. Comparative analysis of haplotype association mapping algorithms. BMC Bioinformatics 2006;7:61.
- Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, et al. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci U S A 2004;101(16):6062-7.
- Pletcher MT, McClurg P, Batalov S, Su AI, Barnes SW, Lagler E, Korstanje R, Wang X, Nusskern D, Bogue MA, et al. Use of a dense single nucleotide polymorphism map for in silico mapping in the mouse. PLoS Biol 2004;2(12):e393.
