The Bioinformatics Research Laboratory (BRL, www.brl.bcm.tmc.edu), directed by Dr. Milosavljevic, develops new experimental and computational methods for discovery through comparative genomics and informatics. BRL aims to enable new discovery methodologies by applying next-generation sequencing and other high-throughput genomic technologies. The technologies, methods, and software systems developed at BRL are applied to study evolution of human genome structure, structural genome changes causing disease, and pathological evolution of the cancer genome.

The Genboree Discovery System (www.genboree.org) is the largest software system developed at BRL in close collaboration with the Human Genome Sequencing Center. Genboree is a turnkey software system for genomic research. Genboree is hosted on the Internet and, as of early 2007, the number of registered users exceeds 600. While it can be configured to support almost any genome-centric discovery process, a number of configurations already exist for specific applications. Current focus is on enabling studies of genome variation, including array CGH studies, PCR-based resequencing, genome resequencing using comparative sequence assembly, genome remapping using paired-end tags and sequences, genome analysis and annotation, multi-genome comparison and pattern discovery via genome self-comparison.

Genboree database and visualization settings, tools, and user roles are configurable to fit the needs of specific discovery processes. Private permanent project-specific databases can be accessed in a controlled way by collaborators via the Internet. Project-specific data is integrated with relevant data from public sources such as genome browsers and genomic databases. Data processing tools are integrated using a plug-in model. Genboree is extensible via flexible data-exchange formats to accommodate project specific tools and processing steps.

Our Positional Hashing method, implemented in the Pash program, enables extremely fast and accurate sequence comparison and pattern discovery by employing low-level parallelism. Pash enables fast and sensitive detection of orthologous regions across mammalian genomes, and fast anchoring of hundreds of millions of short sequences produced by next-generation sequencing technologies. We are further developing the Pash program and employing it in the context of various discovery pipelines.

Our laboratory participates in the pilot stage of the TCGA (The Cancer Genome Atlas) project. We aim to develop comprehensive, rapid, and economical methods for detecting recurrent chromosomal aberrations in cancer using next-generation sequencing technologies. The methods will allow detection of recurrent chromosomal aberrations in hundreds of small (< 1000 cells) specimens at the kilobasepair level of resolution, including small deletions and aberrant joins induced by balanced rearrangements. One of the key objectives is to detect functionally significant recurrent rearrangements in cancer that are positively selected during pathological evolution of cancer cells. Recurrent chromosomal aberrations are best understood in leukemias, lymphomas, and sarcomas. However, recent evidence suggests that carcinomas also contain important recurrent rearrangements, which are not detectable using current methods. We anticipate that the study of highly rearranged genomes such as those found in breast cancer and other carcinomas will particularly benefit from the increase in resolution that we aim to achieve.

Selected publications:

1. Harris RA, Rogers J, Milosavljevic A (2007) Human-specific changes of genome structure detected by genomic triangulation. Science, Apr 13;316(5822):235-7.
   (PubMed) (Journal)

2. Rhesus Macaque Genome Sequencing and Analysis Consortium (2007) Evolutionary and biomedical insights from the rhesus macaque genome. Science, Apr 13;316(5822):222-34.
   (PubMed) (Journal)

3. Milosavljevic A, Harris RA, Sodergren EJ, Jackson AR, Kalafus KJ, Hodgson A, Cree A, Dai W, Csuros M, Zhu B, de Jong PJ, Weinstock GM, Gibbs RA (2005) Pooled Genomic Indexing of Rhesus Macaque. Genome Research , 15:292-301.
   (PubMed) (Journal)

4. Csuros M, Milosavljevic A (2004). Pooled Genomic Indexing (PGI): Analysis and Design of Experiments. J Comput Biol, 11(2): 1001-1021.
   (PubMed) (Journal)

5. Rat Genome Sequencing Consortium (2004). Genome sequence of the Brown Norway rat yields insights into mammalian evolution. Nature, 428: 493-521.
   (PubMed)     (Journal)

6. Kalafus KJ, Jackson AR, and Milosavljevic A (2004). Pash: Efficient Genome-Scale Sequence Anchoring by Positional Hashing. Genome Res. 14: 672-678.
   (PubMed)     (Journal)

7. Milosavljevic A (1999). Discovering Patterns in DNA Sequences by the Algorithmic Significance Method. In Pattern Discovery in Biomolecular Data: Tools, Techniques, and Applications, Wang J.T.L., Shapiro B.A. , Shasha D. (eds.). Oxford: Oxford University Press, p. 3-23.