An algorithm and program for finding sequence specific oligonucleotide probes for species identification

Abstract

Background: The identification of species or species groups with specific oligo-nucleotides as molecular signatures is becoming increasingly popular for bacterial samples. However, it shows also great promise for other small organisms that are taxonomically difficult to tract.

Results: We have devised here an algorithm that aims to find the optimal probes for any given set of sequences. The program requires only a crude alignment of these sequences as input and is optimized for performance to deal also with very large datasets. The algorithm is designed such that the position of mismatches in the probes influences the selection and makes provision of single nucleotide outloops. Program implementations are available for Linux and Windows.

Figure 1

Scheme of the probe finding algorithm. Details are explained in the text.

Figure 2

Scheme of the single-nucleotide outloop problem; asterisks represent mismatches, columns represent matches.

Figure 3

Comparison of specific oligos suggested by ARB and PROBE for Thermotoga maritima, in comparison to the whole SSU database. A) Oligo suggested by ARB, but found to have lower than 70% melting probability in two other species. This was therefore rejected by PROBE because of insufficient specificity. B) Oligo suggested by ARB, but found to have lower than 70% melting probability when outlooping is considered. This was therefore also rejected by PROBE because of insufficient specificity. C) Oligo suggested by both programs, whereby the best outgroup matches have a higher than 70% melting probability.