The Computational Biology Center

Computational Biology & Medical Informatics


The Computational Biology Center embraces activities at Yorktown Heights, with strong affiliations with activities at Almaden and other IBM Research Centers. Computational Biology (CompBio) including bioinformatics is the study of how computer systems can manage, analyze, and simulate the complex structures and processes inherent in living systems. CompBio Research at IBM spans pattern recognition in sequences, structures and processes, the studying of systems ranging from single protein molecules through to complex molecular interactions, and the data analysis, interpretation and reverse-engineering of complex disease-lifestyle-genomic interactions for fuller biological understanding. "CompBio" has a flavor of its own independant of its parents, biology and computer science. Nonetheless, CompBio is inherently a multi- disciplinary field with important applications in biology, chemical physics, materials science, agriculture, chemistry and ultimately nanotechnology.


It is heart warming for any biologist, computational as well as experimental, to be able to see direct benefits of his work to human health and well-being. Hence it is of relevance that there has been a dramatic growth of interest in "personalized medicine" (i.e not a general treatment, but a personally tailored one) based on impacts of the Human Genome Projects, population genomics, digitalization of the patient record, and pharmacogenomics and proteomics. The lack of a personalized medicine up until the near future is currently a major cause of inneffective therapies, pharmaceutical-drug-related injury and death. These new activities, which should alleviate these problems, must be heavilly based on computers. Hence there has also recently been a dramatic increase of interest in adapting the CBC tools for biomedical and clinical applications. This has led to a rich variety of potential customers and colaborators. Some of the software developed at the CBC has been mobilized successfully at an important phsyycian-reseracher site in Canada. There have also been formal IBM Research contracts on Information Based Medicine and Computional Biology in a collaborative project involving the Universities and colleges of the State of Virginia, and similarly with Cambridge University UK. There has been an ongoing study of how the algorithms developed at the Computational Biology Center can be used to analyze the content of patient records, so enabling the path to a rational, computer-and-genomics-based personalized medicine.

The external interest in CBC activity has also been in regard to the IBM protein science initiative and hence link strongly to the Blue Gene petaflop computing initiative. This kind of research also has impact on personalized medicine and the CBC's involvement. A few months ago, working with IBM Research Haifa, a contrived patient record (derived from realistic sources but fully anonymous and processed to prevent effective use and disclosure of data) was used to ship patient clinical data and patient DNA data. A fully automatic process maintained confidentiality, confirmed the correct DNA was spliced into othe correct patient record, annotated the DNA and the resulting protein translation from it, took account of the affects of the patient's genetic difference as affecting that annotation, modeled the patient's unique ("polymorphic") protein from a standard template, and finally selected an appropriate drug from simulated binding studies. This brought together three CBC technologies (computational genomics, automatic sequence annotation, and molecular modeling) and a fourth (proteomics and expression anlysis) will soon be added. This was an information technology proof-of-concept rather than a therapeutic one, but it may have been a small historical first and demonstrated an important concept, namely that in some relatively tractable areas, no quantum leaps in technology are required.

Related Publications  

Alvaro Mateos, JoaquęŽn Dopazo, Ronald Jansen, Yuhai Tu, Mark Gerstein and Gustavo Stolovitzky. Systematic Learning of Gene Functional Classes From DNA Array Expression Data by Using Multilayer Perceptrons. Genome Research.

Laxmi Parida and Alberto Apostolico. Compression and the Wheel of Fortune. Data Compression Conference, Utah. March 2003.

Daniel E. Platt, Concettina Guerra, Giuseppe Zanotti and Isidore Rigoutsos. Global Secondary Structure Packing Angle Bias in Protiens. In Proteins: Structure, Function and Genetics, Press. 2003.

Isidore Rigoutsos, Jiri Novotny, Tien Huynh, Stephen Chin-Bow, Laxmi P. Parida, Daniel E. Platt, David Coleman and Thomas Shenk. In Silico Pattern-based Analysis of the Human Cytomegalovirus (HHV5) Genome. Journal of Virology 77(7):4326-44, April 2003.

Isidore Rigoutsos, Tien Huynh, Aris Floratos, Laxmi Parida and Daniel Platt. Dictionary-driven Protein Annotation.. Nucleic Acids Research 30(17):3901-3916, 2002.

Barry Robson. CLINICAL AND PHARMACOGENOMIC DATA MINING. 1. THE GENERALIZED THEORY OF EXPECTED INFORMATION AND APPLICATION TO THE DEVELOPMENT OF TOOLS. Journal of Proteome Research 2(3):283-302, January 2003.

Recent Accomplishments

Barry Robson, appointed Honorary Professor, St. Matthews

Barry Robson, Professorial Lecturer, Mt Sinai Medical School, NY

Barry Robson, reappointed as Executive Chair, The Dirac Foundation, St. Mary's Hospital, imperial College London.

Isidore Rigoutsos elected Fellow of the Americal Institute for Medical and Biological Engineering

Isidore Rigoutsos invited speaker for Gordon Research Conference on Bioinformatics, Queen's College in Oxford

Isidore Rigoutsos Associate Editor for the journal "Genomics"

Rate this article

 


Image
Structure of the BamHI-DNA complex. Ball and sticks represent the QM subsystem in our model.