Evaluating bacterial gene-finding HMM structures as probabilistic logic programs

Abstract

Motivation: Probabilistic logic programming offers a powerful way to describe and evaluate structured statistical models. To investigate the practicality of probabilistic logic programming for structure learning in bioinformatics, we undertook a simplified bacterial gene-finding benchmark in PRISM, a probabilistic dialect of Prolog.

Results: We evaluate Hidden Markov Model structures for bacterial protein-coding gene potential, including a simple null model structure, three structures based on existing bacterial gene finders and two novel model structures. We test standard versions as well as ADPH length modeling and three-state versions of the five model structures. The models are all represented as probabilistic logic programs and evaluated using the PRISM machine learning system in terms of statistical information criteria and gene-finding prediction accuracy, in two bacterial genomes. Neither of our implementations of the two currently most used model structures are best performing in terms of statistical information criteria or prediction performances, suggesting that better-fitting models might be achievable.

Availability: The source code of all PRISM models, data and additional scripts are freely available for download at: http://github.com/somork/codonhmm.

Supplementary information: Supplementary data are available at Bioinformatics online.

Fig. 1.

Graphical representation of the conditioning schemes of the underlying structure of the models. (a) iid.psm; (b) eco.psm; (c) i3pmc.psm; (d) mc5.psm; (e) aa.psm; and (f) mm.psm. Squares represent the hidden State(S), Previous State(PS) or Next State(S); circles represent emissions (X) or past emissions (P). The dotted arrows are conditional transition probabilities and the full arrows are conditional emission probabilities.

Fig. 2.

E.coli cross-validation ROC curves for the standard models using thresholds over all log odds values in the positive set. Notice that only the area from 0.0–0.1 FPR and 0.9–1.0 TPR is shown.