Effective Alignment of RNA Pseudoknot Structures Using Partition Function Posterior Log-Odds Scores Yang Song1, Lei Hua1, Bruce a Shapiro2 and Jason TL Wang1*

Effective Alignment of RNA Pseudoknot Structures Using Partition Function Posterior Log-Odds Scores Yang Song1, Lei Hua1, Bruce a Shapiro2 and Jason TL Wang1*

Song et al. BMC Bioinformatics (2015) 16:39 DOI 10.1186/s12859-015-0464-9 METHODOLOGY ARTICLE Open Access Effective alignment of RNA pseudoknot structures using partition function posterior log-odds scores Yang Song1, Lei Hua1, Bruce A Shapiro2 and Jason TL Wang1* Abstract Background: RNA pseudoknots play important roles in many biological processes. Previous methods for comparative pseudoknot analysis mainly focus on simultaneous folding and alignment of RNA sequences. Little work has been done to align two known RNA secondary structures with pseudoknots taking into account both sequence and structure information of the two RNAs. Results: In this article we present a novel method for aligning two known RNA secondary structures with pseudoknots. We adopt the partition function methodology to calculate the posterior log-odds scores of the alignments between bases or base pairs of the two RNAs with a dynamic programming algorithm. The posterior log-odds scores are then used to calculate the expected accuracy of an alignment between theRNAs.Thegoalistofindanoptimalalignmentwiththe maximum expected accuracy. We present a heuristic to achieve this goal. The performance of our method is investigated and compared with existing tools for RNA structure alignment. An extension of the method to multiple alignment of pseudoknot structures is also discussed. Conclusions: The method described here has been implemented in a tool named RKalign, which is freely accessible on the Internet. As more and more pseudoknots are revealed, collected and stored in public databases, we anticipate a tool like RKalign will play a significant role in data comparison, annotation, analysis, and retrieval in these databases. Keywords: RNA secondary structure including pseudoknots, Structural alignment, Dynamic programming algorithm Background sequence data (i.e. nucleotides or bases) and structure RNA pseudoknots are formed by pairing bases on data (i.e. base pairs), and produces as output an align- single-stranded loops, such as hairpin and internal loops, ment between the two pseudoknotted RNAs. The struc- with bases outside the loops [1,2]. They are often min- ture data of a pseudoknotted RNA can be obtained from gled with other RNA tertiary motifs [3], and are also the literature or public databases [15-18]. found in non-coding RNAs [4,5]. RNA pseudoknots, RKalign adopts the partition function methodology to with diverse functions [6,7], play important roles in calculate the posterior probabilities or log-odds scores of many biological processes [8,9]; for example, they are re- structural alignments. The idea of using posterior prob- quired for telomerase activity [7], and have been shown abilities to align biomolecules originated from [19,20] to regulate the efficiency of ribosomal frameshifting in where the partition function methodology was employed viruses [10]. to calculate the posterior probabilities of protein se- Analysis and detection of RNA pseudoknots has been quence alignments. Similar techniques were proposed by an active area of research. Many published articles in Do et al. [21] where the authors used hidden Markov this area were focused on pseudoknot alignment [11-14]. models (HMMs) to calculate the posterior probabilities. In this paper, we present a new approach, called RKalign, Will et al. [22] extended the idea of [19-21] to structure- for RNA pseudoknot alignment. RKalign accepts as in- based multiple RNA alignment where the authors put two pseudoknotted RNAs where each RNA has both calculated partition functions inside and outside of sub- sequence pairs on two pseudoknot-free RNAs. Here, we * Correspondence: wangj@njit.edu further extend this idea to pseudoknot alignment. 1Bioinformatics Program, Department of Computer Science, New Jersey Institute of Technology, University Heights, Newark, New Jersey 07102, USA Full list of author information is available at the end of the article © 2015 Song et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Song et al. BMC Bioinformatics (2015) 16:39 Page 2 of 15 Several tools are available for RNA sequence- sequence and 3D structure information. The above structure alignment [23-25]. These tools do not deal methods and tools were mainly designed for aligning with pseudoknots. Mohl et al. [26] proposed a method two RNA tertiary structures by considering their geomet- to perform sequence-structure alignment for RNA ric properties and torsion angles. In contrast, RKalign pseudoknots. The authors set up a pipeline for com- is used to align two RNA secondary structures with bining alignment and prediction of pseudoknots, and pseudoknots. showed experimentally the effectiveness of this pipe- The second group of algorithms is concerned with line in pseudoknot structure annotation. Han et al. aligning two RNA secondary structures without pseudo- [27] decomposed embedded pseudoknots into simple knots. These algorithms employed general edit-distance pseudoknots and aligned them recursively. Yoon [28] alignment [38] or tree matching techniques [39-41]. used a profile-HMM to establish sequence alignment Jiang et al. [42] developed an approximation algorithm constraints, and incorporated these constraints into for aligning a pseudoknot-free structure with a pseu- an algorithm for aligning RNAs with pseudoknots. doknotted structure. Our work differs from Jiang Wong et al. [29] identified the pseudoknot type of a et al.’s work in that we focus on the alignment of two given structure and developed dynamic programming pseudoknotted structures. Furthermore, we use the algorithms for structural alignments of different pseu- partition function methodology whereas Jiang et al. doknot types. Huang et al. [4] applied a tree decom- adopted a general edit-distance approach to the struc- position algorithm to search for non-coding RNA tural alignment. pseudoknot structures in genomes. The method that is most closely related to ours is an The above methods were concerned with aligning a option offered by the CARNA tool [14]. Like RKalign, pseudoknot structure with a sequence or genome. this option is able to accept two known RNA secondary Through the alignment, the sequence is folded and its structures with pseudoknots, and produce an alignment structure is predicted. Xu et al. [11] presented a differ- between the two RNA structures. This option employs ent method, called RNA Sampler, which can simultan- constraint programming techniques with a branch and eously fold and align two or multiple RNA sequences bound scheme. It gradually refines solutions until the considering pseudoknots without known structures. best solution is found. To understand the relative per- Similar techniques were implemented in DAFS [12] and formance of the two tools, we perform extensive experi- SimulFold [13]. Additional methods can be found in the ments to compare RKalign with CARNA using different CompaRNA web server [30]. In contrast to these datasets. methods, which perform alignment and folding at the same time, RKalign aims to align two known RNA pseu- Methods doknot structures where the structures are obtained In this section, we present algorithmic details of RKalign. from existing databases [15-17]. As more pseudoknot To align two RNA pseudoknot structures A and B,we structures become available in these databases, a tool adopt the partition function methodology to calculate like RKalign will be useful in performing data analysis in the posterior probabilities or log-odds scores of the the repositories. alignments between bases or base pairs in A and B re- There are two groups of algorithms which are also cap- spectively. After calculating the posterior log-odds scores, able of aligning two known RNA structures. The first group we then compute the expected accuracy of an alignment is concerned with aligning two RNA three-dimensional between structure A and structure B.Thegoalistofind (3D) structures, possibly containing pseudoknots. Ferre an optimal alignment between A and B where the align- et al. [31] presented a dynamic programming algorithm by ment has the maximum expected accuracy. We will taking into account nucleotide, dihedral angle and base- present a heuristic to achieve this goal. pairing similarities. Capriotti and Marti-Renom [32] devel- oped a program to align two RNA 3D structures based on a unit-vector root-mean-square approach. Chang et al. [33] Definitions and notation and Wang et al. [34] employed a structural alphabet of Suppose (i, j) is a base pair of pseudoknot structure different nucleotide conformations to align RNA 3D A and (p, q) is a base pair of pseudoknot structure structures. Hoksza and Svozil [35] developed a pairwise B.Weusescore((i, j), (p, q)) to represent the score comparison method based on 3D similarity of general- of aligning (i, j)with(p, q)wherethescoreisob- ized secondary structure units. Rahrig et al. [36] pre- tained from the log-odds RIBOSUM matrix [43]. sented the R3D Align tool for performing global The use of this scoring matrix permits RKalign to pairwise alignment of RNA 3D structures using local determine the similarity between pseudoknot struc- superpositions.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    15 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us