Rapid large-scale oligonucleotide selection for microarrays

We present the first algorithm that selects oligonucleotide probes (e.g. 25-mers) for microarray experiments on a large scale. For example, oligos for human genes can be found within 50 hours. This becomes possible by using the longest common substring as a specificity measure for candidate oligos. We present an algorithm based on a suffix array with additional information that is efficient both in terms of memory usage and running time to rank all candidate oligos according to their specificity. We also introduce the concept of master sequences to describe the sequences from which oligos are to be selected. Constraints such as oligo length, melting temperature, and self-complementarity are incorporated in the master sequence at a preprocessing stage and thus kept separate from the main selection problem. As a result, custom oligos can now be designed for any sequenced genome, just as the technology for on-site chip synthesis is becoming increasingly mature.