Sequence data analysis and preprocessing for oligo probe design in microbial genomes

Abstract

A good oligo probe design in DNA microarray experiments is crucial to obtain the better results of gene expression analysis. However, sequence data from a very large microbial genome or pan-genome will produce a reduced number of oligos and affect the design quality if processed by a probe designer. Gene redundancies and discrepancies across resources of the same species or strain and their sequence similarity and homology are responsible for the poor quantity of oligos designed. We addressed these issues and problems with sequences and introduced the concept of open reading frame (ORF) sequence segmentation from which quality oligos can be selected. Analysis and pre-processing of sequence data were performed using our Perl-based pipeline ORF-Purger 2.0. ORFs were purged of redundancy, discrepancy, invalidity, overlapping, similarity and, optionally, homology, such that the quantity and quality of oligos to be designed were drastically improved. Probe integrity was proposed as the first probe selection criterion since the fully physical availability of all possible probes corresponding to their targets in a nucleic acid sample is necessary for a best probe design.