Title: Assistant Professor, Center for Computational Molecular Biology, Brown University
Education: PhD, University of San Diego, 2002
Recommended by: Bud Mishra
Ben Raphael wasn't looking for an opportunity to apply his algorithmic and computational know-how to cancer research. It found him, and he's grateful.
Back in 2003, Raphael was busy examining rearrangements of the mouse and human genome in an evolutionary context when a group from the University of California, San Francisco, approached him with an unexpectedly scrambled cancer genome. “It actually happened somewhat serendipitously,” remembers Raphael. The researchers, having analyzed the genome using clone end sequencing, were surprised with the results, having based their expectations on earlier cytogenetic studies. So Raphael began helping to develop computational techniques to analyze the mixed-up data.
“It's been quite a lucky thing for me to be part of such a great collaboration. Starting this project really set me on this direction, which is now becoming a pretty hot area,” he says. It's now known that cancer can stem from a host of mutations, including single base pair changes, chromosomal inversions, duplications, translocations, and deletions — all changes that can alter the structure or regulation of genes. “In leukemia, there are several known fusion genes that are created by translocations that merge genes from two different chromosomes,” he says. This not only applies to leukemia, but to blood cancers, lymphomas, and prostate and ovarian cancer as well.
“When you look at advanced-stage tumors, the genome tends to be extensively rearranged,” Raphael says. His initial work on the tumor data produced the first high-resolution reconstruction of a tumor genome. He has continued to develop algorithms for genome rearrangement analysis using clone end sequence profiling. By understanding the large-scale changes that take place in tumor genomes due to extensive rearrangement, he hopes to contribute to the development of targeted cancer treatments.
With many sequencing centers now shifting their focus toward cancer and tumor sequencing, Raphael believes that genomic cancer research will receive quite a boost. “I got involved a couple of years ago, slightly before it was becoming a big thing,” he says. “And I think it's really kind of at this tipping point where there's going to be a lot of exciting things coming out in the next few years.”
Raphael's other main focus is producing algorithms to address DNA and protein sequencing problems, such as multiple sequence alignment and motif finding. He recently helped to develop a new method for multiple sequence alignment called A-Bruijn Alignment. According to Raphael, ABA is an improvement upon similar algorithms for multiple sequence alignment because it permits the alignment of sequences with self-similar or shuffled subsequences.
Raphael is eager for the day when people can walk into the clinic, have their tumors sequenced, and get information on the important mutations that point to a specific treatment program within minutes. But until that time, he is looking forward to tackling the computational problems associated with sequencing tumor genomes. “When you've got something like a cancer genome where there's extensive duplication, it's even more challenging,” he says.
Publications of note
Early this year, Raphael and his colleagues published a paper in Genome Research called “Decoding the fine-scale structure of a breast cancer genome and transcriptome.” The study demonstrated advantages of end sequence profiling to map the rearrangements of tumor genomes using the MCF-7 breast cancer cell line; those include the ability to generate tumor-specific reagents for in vitro and in vivo studies as well as detection of rearrangement and copy number changes.