Edison T. LIU
Executive Director, Genome Institute of Singapore
Professor of Medicine, National University of Singapore

Dr Liu is the Executive Director, Genome Institute of Singapore (Agency for Science, Technology, and Research) and Professor of Medicine, National University of Singapore. He received his degrees (B.S. and M.D.) at Stanford University further oncology and molecular biology training at Stanford and at UCSF. Prior to starting the GIS in 2001, Dr. Liu was the Scientific Director for the Division of Clinical Sciences, National Cancer Institute (USA). Dr. Liu’s scientific research has focused on the functional genomics of human cancers, particularly breast cancer, investigating the dynamics of gene regulation on a genome scale that modulates cancer biology. He has authored over 260 scientific papers, reviews and books. He is the recipient of a number of awards: Leukemia Society Scholar, 1991-96; Member, the American Society of Clinical Investigations (elected 1995), the Brinker International Award for basic science research in Breast Cancer, 1996; the Rosenthal Award from the American Association for Cancer Research, 2000; the President’s Public Service Medal for helping Singapore resolve the SARS crisis, and a Doctor of Medicine Sciences honoris causa,Queen’s University, Belfast, 2007. He was elected as a Foreign Associate Member of the European Molecular Biology Organization (EMBO, 2008), and serving his second term as the elected President of the Human Genome Organization (HUGO). Dr. Liu is also the current Chairman of the Board for the Singapore Health Sciences Authority, health regulatory and blood banking agency for Singapore.

Abstract Title: Cancer Genomes in Personalized Medicine

Assessing and understanding genetic complexity in cancer will be a key hurdle as we move towards tailoring specific therapies for cancer. DNA paired end ditag (DNA PET) is a cloning strategy where paired ends of genomic fragments of specific length (1, 5, 10 kb) are cloned and isolated for sequencing using next generation technologies. We sequenced a total 111 million pair end ditags (PETs) from three MCF-7 libraries which provided 136X coverage of the cancer genome and identified 9.6 million PETS that assembled into ~2,400 clusters where the mapping of the PETs showed significant deviation from the reference genome. These discordant PETs (D-PETs) represent loci that are candidates for genomic rearrangements. Gene amplification could be precisely determined by PET counts at any genomic coordinate when compared to the normalized genomic background, and were found to be superior to aCGH in dynamic range, coverage, and precision. Moreover, the reproducibility was excellent (R = 0.99). Upon detailed analysis and validation, D-PETs correctly mapped all deletions, insertions, inversions, translocations in MCF7 and provided the breakpoint coordinates with ~150 bp resolution. When the potential for fusions genes was assessed, we found approximately 753 intra-chromosomal events that may generate a potential fusion gene, and 31 inter-chromosomal rearrangements what might potential fusion genes. There was high complexity of the rearrangements present with a number of translocations that were within or at the borders of segments of gene amplification. We had previously reported a detailed analysis of full length cDNAs in MCF-7 (Ruan Y, et al. Fusion transcripts and transcribed retrotransposed loci discovered through comprehensive transcriptome analysis using Paired-End diTags (PETs). Genome Res. 2007 Jun;17(6):828-38.). The intersection of the two datasets showed overlap between the D-PETs between the two platforms. Intriguingly, a significant number of the fusion transcripts arose from tandem duplications.

Back to Speakers