The function of a non-protein-coding RNA is often determined by its structure. Since experimental determination of RNA structure is time-consuming and expensive, its computational prediction is of great interest, and efficient solutions based on thermodynamic parameters are known. Frequently, however, the predicted minimum free energy structures are not the native ones, leading to the necessity of generating suboptimal solutions. While this can be accomplished by a number of programs, the user is often confronted with large outputs of similar structures, although he or she is interested in structures with more fundamentaldifferences,or, inotherwords, with different abstract shapes. Here, we formalize the concept of abstract shapes and introduce their efficient computation. Each shape of an RNA molecule comprises a class of similar structures and has a representative structure of minimal free energy within the class. Shape analysis is implemented in the program RNAshapes. We applied RNAshapes to the prediction of optimal and suboptimal abstract shapes of severalRNAs.For a given energy range, the number of shapes is considerably smaller than the number of structures, and in all cases, the native structures were among the top shape representatives. This demonstrates that the researcher can quickly focus on the structures of interest, without processing up to thousands of near-optimal solutions. We complement this study with a large-scale analysis of the growth behaviour of structure and shape spaces. RNAshapes is available for download and as an online version on the Bielefeld Bioinformatics Server.

Abstract not found

For modern biology, precise genome annotations are of prime importance, as they allow the accurate definition of genic regions. We employ state-of-the-art machine learning methods to assay and improve the accuracy of the genome annotation of the nematode Caenorhabditis elegans. The proposed machine learning system is trained to recognize exons and introns on the unspliced mRNA, utilizing recent advances in support vector machines and label sequence learning. In 87 % (coding and untranslated regions) and 95 % (coding regions only) of all genes tested in several out-ofsample evaluations, our method correctly identified all exons and introns. Notably, only 37 % and 50%, respectively, of the presently unconfirmed genes in the C. elegans genome annotation agree with our predictions, thus we hypothesize that a sizable fraction of those genes are not correctly annotated. A retrospective evaluation of the Wormbase WS120 annotation [1] of C. elegans reveals that splice form predictions on unconfirmed genes in WS120 are inaccurate in about 18 % of the considered cases, while our predictions deviate from the truth only in 10%–13%. We experimentally analyzed 20 controversial genes on which our system and the annotation disagree, confirming the superiority of our predictions. While our method correctly predicted 75 % of those cases, the standard annotation was never completely correct. The accuracy of our system is further corroborated by a comparison with two other recently proposed systems that can be used for splice form prediction: SNAP and ExonHunter. We conclude that the genome annotation of C.

Many combinatorial optimization problems in biosequence analysis are solved via dynamic programming. To increase programming productivity and program reliability, a domain specific language embedded in Haskell has been suggested. We point out several shortcomings of this approach, and report on some challenges in the (ongoing) project of migrating this domain specific language from its host language to a directly compiled implementation. Most of these challenges are domain specific optimizations, which not only improve significant constant factors of runtime and space requirements, but also affect asymptotic efficiency. We report on our solutions to some of these problems, and point out others that are still open.

Recombination is an important event in the evolution of HIV. It affects the global spread of the pandemic as well as evolutionary escape from host immune response and from drug therapy within single patients. Comprehensive computational methods are needed for detecting recombinant sequences in large databases, and for inferring the parental sequences. We present a hidden Markov model to annotate a query sequence as a recombinant of a given set of aligned sequences. Parametric inference is used to determine all optimal annotations for all parameters of the model. We show that the inferred annotations recover most features of established hand-curated annotations. Thus, parametric analysis of the hidden Markov model is feasible for HIV full-length genomes, and it improves the detection and annotation of recombinant forms. All computational results, reference alignments, and C++ source code are available at

Thermodynamic RNA secondary structure prediction is an important recipe for the latest generation of functional non-coding RNA finding tools. However, the predicted energy is not strong enough by itself to distinguish a single functional non-coding RNA from other RNA. Here, we analyze how well an RNA molecule folds into a particular structural class with a restricted folding algorithm called Thermodynamic Matcher (TDM). We compare this energy value to that of randomized sequences. We construct and apply TDMs for the non-coding RNA families RNA I and hammerhead ribozyme type III and our results show that using TDMs rather than universal minimum free energy folding allows for highly significant predictions. 1.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License

Abstract not found

the thermodynamic model for RNA secondary structure prediction

For modern biology, precise genome annotations are of prime importance, as they allow the accurate definition of genic regions. We employ state-of-the-art machine learning methods to assay and improve the accuracy of the genome annotation of the nematode Caenorhabditis elegans. The proposed machine learning system is trained to recognize exons and introns on the unspliced mRNA, utilizing recent advances in support vector machines and label sequence learning. In 87 % (coding and untranslated regions) and 95 % (coding regions only) of all genes tested in several out-ofsample evaluations, our method correctly identified all exons and introns. Notably, only 37 % and 50%, respectively, of the presently unconfirmed genes in the C. elegans genome annotation agree with our predictions, thus we hypothesize that a sizable fraction of those genes are not correctly annotated. A retrospective evaluation of the Wormbase WS120 annotation [1] of C. elegans reveals that splice form predictions on unconfirmed genes in WS120 are inaccurate in about 18 % of the considered cases, while our predictions deviate from the truth only in 10%–13%. We experimentally analyzed 20 controversial genes on which our system and the annotation disagree, confirming the superiority of our predictions. While our method correctly predicted 75 % of those cases, the standard annotation was never completely correct. The accuracy of our system is further corroborated by a comparison with two other recently proposed systems that can be used for splice form prediction: SNAP and ExonHunter. We conclude that the genome annotation of C.