Gene Chip Design Technology Research

Microarrays are high density probe arrays, which are composed of short DNA strands, so called probes, arranged on the solid support media according to the definite order. Probes and special nucleotides will hybridize by nucleic acid hybridize principle under appropriate condition, Due to this property, microarrys can be used to detect nucleotide sequences in the given samples. Microarrays can analysis tens of thousands of pieces of information in genome simultaneously and have a great value of basic research and obvious industrialized foreground,The selection of probes and optimal masking are two problems arising in the design of microarrays.The task of probe selection is to select specific probes for aim-genes in the given genome, and the probes selected should have the ability of distinguishing aim-genes from nonaim-genes, in other words, the probes on chip should be hybridized only with the aim-gene sequences, and not with any other sequences in the genome. It is difficult to select proper probes because there are too much data in genome and too many intended probes. In this paper, a whole algorithm of probe selection is given to reduce the calculation in the process of probe selection by using a new rule, markov probability rule. In the process of probe selection, the markov probability rule is used to forecast the specialness of the probes and the probes, which are more likely to be special ones, are selected to be computed to reduce the number of probes which should be evaluated by searching in the genome, Moreover, the data structure of suffix array is adapted to reduce the calculation. When the algorithm is applied to the test data of HIV-1 subtypes, the result shows that the number of probes, which require calculating in the process of selecting probes, is reduced to the tenth of candidate probes, As can prove the algorithm is effective.Masks play an important role in the process of synthesizing probes for microarray. Because fabricating masks is laborious and costly, in the process of synthesizing probes for microarray, masks should be designed optimally so as to reduce their amount without changing the synthesizing result. In this paper, a look-ahead search algorithm is proposed to optimize the design of masks. Applied to test data, the algorithm reduces more than 20% masks compared to the traditional method.