AND MICROARRAYS ARE A COMPLEX TECHNOLOGY USED IN MOLECULAR BIOLOGY AND MEDICINE TO SIMULTANEOUSLY MONITOR CHANGES IN THE EXPRESSION LEVELS OF MULTIPLE GENES. TODAY, MICROARRAYS HAVE MULTIPLE APPLICATIONS: TO IDENTIFY GENES WHOSE EXPRESSION CHANGES IN RESPONSE TO PATHOGENS OR OTHER ORGANISMS BY COMPARING GENE EXPRESSION IN INFECTED CELLS/TISSUES WITH THOSE IN UNINFECTED CELLS/TISSUES; TO COMPARE GENE EXPRESSION IN TUMOUR PROFILES FROM CANCER PATIENTS; OR GENE EXPRESSION IN HEART DISEASE PROFILES. THEY ALSO HAVE IMPORTANT APPLICATIONS IN CLINICAL AND PHARMACEUTICAL RESEARCH. THE OBJECTIVE OF THIS PROJECT IS TO BUILD A TECHNIQUE FOR SEGMENTATION OF DNA ARRAYS THAT WORKS BETTER THAN THE METHODS ALREADY USED. WE HAVE ALREADY PRESENTED A PRELIMINARY SEGMENTATION TECHNIQUE IN A RECENT ISI PUBLICATION AND WE AIM TO IMPROVE THIS TECHNIQUE AND ALSO TO COMPARE IT WITH EXISTING SEGMENTATION TECHNIQUES, HIGHLIGHTING THE DIFFERENCES. THIS WILL REQUIRE SIGNIFICANT STATISTICAL ANALYSIS BASED ON THE RESULTS GENERATED BY DIFFERENT METHODS, AS WELL AS THE IMPLEMENTATION OF THE METHODS, BACKGROUND IN BIOLOGY AND, OF COURSE, THE NECESSARY BIONFORMATICS TOOLS. IN SHORT, THE PROJECT WILL REQUIRE EXPERTS FROM STATISTICS, MATHEMATICS, COMPUTER SCIENCE AND BIOLOGY IN A TRULY INTERDISCIPLINARY COLLABORATION.

IN THE LONG TERM, WE FORESEE THAT THE PROJECT COULD LEAD TO THE DEVELOPMENT OF A SEGMENTATION METHOD THAT COULD PROVIDE QUANTITATIVE RESULTS FROM DNA ARRAY EXPERIMENTS (AT THIS MOMENT WE ARE RATHER RECEIVING QUALITATIVE SIGNALS OF THE GENE-DISEASE RELATIONSHIP). WE EXPECT THAT THE RESOLUTION OF THE IMAGES OBTAINED FROM THE ADN ARRAY EXPERIMENTS WILL INCREASE AND CONSEQUENTLY WILL LEAD THE PROPOSED ALGORITHMS TO A SIGNIFICANT IMPROVEMENT, UNLIKE THE AFFYMETRIX SEGMENTATION METHOD. THIS WILL LEAD TO QUANTITATIVE RESULTS, WHICH CAN HAVE A SIGNIFICANT IMPACT ON CELLULAR PROCESSES.