MoTeX-II: structured MoTif eXtraction from large-scale datasets

Abstract

BackgroundIdentifying repeated factors that occur in a string of letters or common factors that occur in a set of strings represents an important task in computer science and biology. Such patterns are called motifs, and the process of identifying them is called motif extraction. In biology, motif extraction constitutes a fundamental step in understanding regulation of gene expression. State-of-the-art tools for motif extraction have their own constraints. Most of these tools are only designed for single motif extraction; structured motifs additionally allow for distance intervals between their single motif components. Moreover, motif extraction from large-scale datasets—for instance, large-scale ChIP-Seq datasets—cannot be performed by current tools. Other constraints include high time and/or space complexity for identifying long motifs with higher error thresholds.ResultsIn this article, we introduce , a word-based high-performance computing tool for structured MoTif eXtraction from large-scale datasets. Similar to its predecessor for single motif extraction, it uses state-of-the-art algorithms for solving the fixed-length approximate string matching problem. It produces similar and partially identical results to state-of-the-art tools for structured motif extraction with respect to accuracy as quantified by statistical significance measures. Moreover, we show that it matches or outperforms these tools in terms of runtime efficiency by merging single motif occurrences efficiently. comes in three flavors: a standard CPU version; an OpenMP-based version; and an MPI-based version. For instance, the MPI-based version of requires only a couple of hours to process all human genes for structured motif extraction on 1056 processors, while current sequential tools require more than a week for this task. Finally, we show that is successful in extracting known composite transcription factor binding sites from real datasets.ConclusionsUse of in biological frameworks may enable deriving reliable and important information since real full-length datasets can now be processed with almost any set of input parameters for both single and structured motif extraction in a reasonable amount of time. The open-source code of is freely available at http://www.inf.kcl.ac.uk/research/projects/motex/.

libFLASM: a software library for fixed-length approximate string matching

“…libFLASM was incorporated into MoTeX-II [11] (the successor of MoTeX [9]), a state-of-the-art tool to identify single and structured motifs. MoTeX-II uses the library to find occurrences, under a pre-specified distance model, of each factor of length ℓ of every sequence in the input sequences.…”

“…Specifically, the problem can be solved in time and space under both distance models. FLASM is well-suited to solving numerous problems in computational molecular biology, including, for instance, multiple circular sequence alignment (MCSA) [8] and motif extraction [9–11]. FLASM is also very useful for solving problems outside the confines of computational biology, such as approximate circular string matching [12] as well as implementing the Chang and Marr index [13].…”

“…Such conserved structured motifs may serve as potential candidates for transcription factor binding sites for a composite regulatory protein [30]. A few tools exist to tackle the ME problem for structured motifs [11, 31, 32]. …”

“…Theorems 1 and 2 can be applied to find occurrences of individual motifs m 1 , m 2 ,…, m β , where β is the number of single motifs within a structured motif; and then merge these occurrences efficiently to form structured motifs. The same validity assumption holds for structured motifs (see [11], for details).…”

An Entropy-Based Position Projection Algorithm for Motif Discovery

“…These algorithms can be divided into two categories: One is exact algorithms, which use consensus sequences to represent motifs [3]. Recent exact algorithms mainly concentrate on pattern-driven algorithms [4–8]. They scan all sequence patterns of length l with an initial search space of O(4 l ) and report all possible solves.…”