A new algorithm for detecting low-complexity regions in protein sequences

MOTIVATION: Pair-wise alignment of protein sequences and local similarity searches produce many false positives because of compositionally biased regions, also called low-complexity regions (LCRs), of amino acid residues. Masking and filtering such regions significantly improves the reliability of homology searches and, consequently, functional predictions. Most of the available algorithms are based on a statistical approach. We wished to investigate the structural properties of LCRs in biological sequences and develop an algorithm for filtering them. RESULTS: We present an algorithm for detecting and masking LCRs in protein sequences to improve the quality of database searches. We developed the algorithm based on the complexity analysis of subsequences delimited by a pair of identical, repeating subsequences. Given a protein sequence, the algorithm first computes the suffix tree of the sequence. It then collects repeating subsequences from the tree. Finally, the algorithm iteratively tests whether each subsequence delimited by a pair of repeating subsequences meets a given criteria. Test results with 1000 proteins from 20 families in Pfam show that the repeating subsequences are a good indicator for the low-complexity regions, and the algorithm based on such structural information strongly compete with others. AVAILABILITY: http://bioinfo.knu.ac.kr/research/CARD/ CONTACT: swshin@bioinfo.knu.ac.kr     

CLAGen: a tool for clustering and annotating gene sequences using a suffix tree algorithm

Most multiple gene sequence alignment methods rely on conventions regarding the score of a multiple alignment in pairwise fashion. Therefore, as the number of sequences increases, the runtime of sequencing expands exponentially. In order to solve the problem, this paper presents a multiple sequence alignment method using a linear-time suffix tree algorithm to cluster similar sequences at one time without pairwise alignment. After searching for common subsequences, cross-matching common subsequences were generated, and sometimes inexact matching was found. So, a procedure aimed at masking the inexact cross-matching pairs was suggested here. In addition, BLAST was combined with a clustering tool in order to annotate the clusters generated by suffix tree clustering. The proposed method for clustering and annotating genes consists of the following steps: (1) construction of a suffix tree; (2) searching and overlapping common subsequences; (3) grouping subsequence pairs; (4) masking cross-matching pairs; (5) clustering gene sequences; (6) annotating gene clusters by the BLAST search. The performance of the proposed system, CLAGen, was successfully evaluated with 42 gene sequences in a TCA cycle (a citrate cycle) of bacteria. The system generated 11 clusters and found the longest subsequences of each cluster, which are biologically significant.     

Conserved key amino acid positions (CKAAPs) derived from the analysis of common substructures in proteins

An all-against-all protein structure comparison using the Combinatorial Extension (CE) algorithm applied to a representative set of PDB structures revealed a gallery of common substructures in proteins (http://cl.sdsc.edu/ce.html). These substructures represent commonly identified folds, domains, or components thereof. Most of the subsequences forming these similar substructures have no significant sequence similarity. We present a method to identify conserved amino acid positions and residue-dependent property clusters within these subsequences starting with structure alignments. Each of the subsequences is aligned to its homologues in SWALL, a nonredundant protein sequence database. The most similar sequences are purged into a common frequency matrix, and weighted homologues of each one of the subsequences are used in scoring for conserved key amino acid positions (CKAAPs). We have set the top 20% of the high-scoring positions in each substructure to be CKAAPs. It is hypothesized that CKAAPs may be responsible for the common folding patterns in either a local or global view of the protein-folding pathway. Where a significant number of structures exist, CKAAPs have also been identified in structure alignments of complete polypeptide chains from the same protein family or superfamily. Evidence to support the presence of CKAAPs comes from other computational approaches and experimental studies of mutation and protein-folding experiments, notably the Paracelsus challenge. Finally, the structural environment of CKAAPs versus non-CKAAPs is examined for solvent accessibility, hydrogen bonding, and secondary structure. The identification of CKAAPs has important implications for protein engineering, fold recognition, modeling, and structure prediction studies and is dependent on the availability of structures and an accurate structure alignment methodology. Proteins 2001;42:148-163.     

Bayesian adaptive sequence alignment algorithms

The selection of a scoring matrix and gap penalty parameters continues to be an important problem in sequence alignment. We describe here an algorithm, the 'Bayes block aligner, which bypasses this requirement. Instead of requiring a fixed set of parameter settings, this algorithm returns the Bayesian posterior probability for the number of gaps and for the scoring matrices in any series of interest. Furthermore, instead of returning the single best alignment for the chosen parameter settings, this algorithm returns the posterior distribution of all alignments considering the full range of gapping and scoring matrices selected, weighing each in proportion to its probability based on the data. We compared the Bayes aligner with the popular Smith-Waterman algorithm with parameter settings from the literature which had been optimized for the identification of structural neighbors, and found that the Bayes aligner correctly identified more structural neighbors. In a detailed examination of the alignment of a pair of kinase and a pair of GTPase sequences, we illustrate the algorithm's potential to identify subsequences that are conserved to different degrees. In addition, this example shows that the Bayes aligner returns an alignment-free assessment of the distance between a pair of sequences.      

A proposal on metrics for identification using nucleic acid sequences

Abstract The need is stressed for attempts to be made to permit diagnostic nucleic acid sequences to be used in a quantitative manner. Sequence differences or binding values should be converted to a distance measure and from this an ultrametric tree should be constructed. A single quantitative determination can yield considerable information about the likely identity of an unknown microorganism when the distance obtained from the sequence is compared with the tree. The concept is illustrated by hypothetical species and genus subsequences, and it is suitable both for successive use of hierarchical subsequences and for automated identification. It is pointed out that entirely specific subsequences for higher taxa may be difficult to discover. These principles will be useful for the future design of diagnostic sequences, including possible application to DNA-DNA pairing.     

Comparison of RNA sequences from widely divergent species

Summary Tests that take into account the effects of gaps have been applied to 5S ribosomal RNA sequences from the bacteria,E. coli andP. fluorescens, and from KB carcinoma cells. The 5S RNAs from KB andP. fluorescens, when compared to that ofE. coli are shown to be more similar than random sequences of the same composition. Intrasequential analyses of 5S RNAs give some evidence for partial gene duplication or repetitive subsequences, but the proposed duplication of Brownlee, Sanger and Barrell (1968) is not supported by our data.     

Estimation of P-values for global alignments of protein sequences.

MOTIVATION: The global alignment of protein sequence pairs is often used in the classification and analysis of full-length sequences. The calculation of a Z-score for the comparison gives a length and composition corrected measure of the similarity between the sequences. However, the Z-score alone, does not indicate the likely biological significance of the similarity. In this paper, all pairs of domains from 250 sequences belonging to different SCOP folds were aligned and Z-scores calculated. The distribution of Z-scores was fitted with a peak distribution from which the probability of obtaining a given Z-score from the global alignment of two protein sequences of unrelated fold was calculated. A similar analysis was applied to subsequence pairs found by the Smith-Waterman algorithm. These analyses allow the probability that two protein sequences share the same fold to be estimated by global sequence alignment. RESULTS: The relationship between Z-score and probability varied little over the matrix/gap penalty combinations examined. However, an average shift of +4.7 was observed for Z-scores derived from global alignment of locally-aligned subsequences compared to global alignment of the full-length sequences. This shift was shown to be the result of pre-selection by local alignment, rather than any structural similarity in the subsequences. The search ability of both methods was benchmarked against the SCOP superfamily classification and showed that global alignment Z-scores generated from the entire sequence are as effective as SSEARCH at low error rates and more effective at higher error rates. However, global alignment Z-scores generated from the best locally-aligned subsequence were significantly less effective than SSEARCH. The method of estimating statistical significance described here was shown to give similar values to SSEARCH and BLAST, providing confidence in the significance estimation. AVAILABILITY: Software to apply the statistics to global alignments is available from http://barton.ebi.ac.uk. CONTACT: geoff@ebi.ac.uk     

Iterative stepwise discriminant analysis: a meta-algorithm for detecting quantitative sequence motifs.

An algorithm is presented for detecting a quantitative pattern in peptide fragments that bind class II major histocompatibility complex (MHC) molecules. It is referred to as a meta-algorithm because it requires successive applications of Stepwise Discriminate Analysis (SDA). On every iteration the best subsequence candidates are selected from sequences known to bind class II MHC molecules. When SDA compares probable binding subsequences with subsequences known not to bind class II MHC molecules, a quantitative model emerges that is capable of classifying subsequences as binding or non-binding. In an iterative manner, the resultant model is utilized as a criterion for selecting probable binding subsequence candidates. The procedure is repeated until models converge. In the illustrated examples, the final models correctly classify over 95% of the peptides in a database of peptides whose binding affinity for HLA-DR1 is known. The final model can then be used to predict the binding affinity of peptides that have not yet been laboratory tested.     

Popitam: towards new heuristic strategies to improve protein identification from tandem mass spectrometry data

In recent years, proteomics research has gained importance due to increasingly powerful techniques in protein purification, mass spectrometry and identification, and due to the development of extensive protein and DNA databases from various organisms. Nevertheless, current identification methods from spectrometric data have difficulties in handling modifications or mutations in the source peptide. Moreover, they have low performance when run on large databases (such as genomic databases), or with low quality data, for example due to bad calibration or low fragmentation of the source peptide. We present a new algorithm dedicated to automated protein identification from tandem mass spectrometry (MS/MS) data by searching a peptide sequence database. Our identification approach shows promising properties for solving the specific difficulties enumerated above. It consists of matching theoretical peptide sequences issued from a database with a structured representation of the source MS/MS spectrum. The representation is similar to the spectrum graphs commonly used by de novo sequencing software. The identification process involves the parsing of the graph in order to emphasize relevant sections for each theoretical sequence, and leads to a list of peptides ranked by a correlation score. The parsing of the graph, which can be a highly combinatorial task, is performed by a bio-inspired algorithm called Ant Colony Optimization algorithm.     

Comparative ab initio prediction of gene structures using pair HMMs

We present a novel comparative method for the ab initio prediction of protein coding genes in eukaryotic genomes. The method simultaneously predicts the gene structures of two un-annotated input DNA sequences which are homologous to each other and retrieves the subsequences which are conserved between the two DNA sequences. It is capable of predicting partial, complete and multiple genes and can align pairs of genes which differ by events of exon-fusion or exon-splitting. The method employs a probabilistic pair hidden Markov model. We generate annotations using our model with two different algorithms: the Viterbi algorithm in its linear memory implementation and a new heuristic algorithm, called the stepping stone, for which both memory and time requirements scale linearly with the sequence length. We have implemented the model in a computer program called DOUBLESCAN. In this article, we introduce the method and confirm the validity of the approach on a test set of 80 pairs of orthologous DNA sequences from mouse and human. More information can be found at: http://www.sanger.ac.uk/Software/analysis/doublescan/     

LaTcOm: a web server for visualizing rare codon clusters in coding sequences

We present LaTcOm, a new web tool, which offers several alternative methods for 'rare codon cluster' (RCC) identification from a single and simple graphical user interface. In the current version, three RCC detection schemes are implemented: the recently described %MinMax algorithm and a simplified sliding window approach, along with a novel modification of a linear-time algorithm for the detection of maximally scoring subsequences tailored to the RCC detection problem. Among a number of user tunable parameters, several codon-based scales relevant for RCC detection are available, including tRNA abundance values from Escherichia coli and several codon usage tables from a selection of genomes. Furthermore, useful scale transformations may be performed upon user request (e.g. linear, sigmoid). Users may choose to visualize RCC positions within the submitted sequences either with graphical representations or in textual form for further processing. AVAILABILITY: LaTcOm is freely available online at the URL http://troodos.biol.ucy.ac.cy/latcom.html.     

From analysis of protein structural alignments toward a novel approach to align protein sequences

Alignment of protein sequences is a key step in most computational methods for prediction of protein function and homology-based modeling of three-dimensional (3D)-structure. We investigated correspondence between "gold standard" alignments of 3D protein structures and the sequence alignments produced by the Smith-Waterman algorithm, currently the most sensitive method for pair-wise alignment of sequences. The results of this analysis enabled development of a novel method to align a pair of protein sequences. The comparison of the Smith-Waterman and structure alignments focused on their inner structure and especially on the continuous ungapped alignment segments, "islands" between gaps. Approximately one third of the islands in the gold standard alignments have negative or low positive score, and their recognition is below the sensitivity limit of the Smith-Waterman algorithm. From the alignment accuracy perspective, the time spent by the algorithm while working in these unalignable regions is unnecessary. We considered features of the standard similarity scoring function responsible for this phenomenon and suggested an alternative hierarchical algorithm, which explicitly addresses high scoring regions. This algorithm is considerably faster than the Smith-Waterman algorithm, whereas resulting alignments are in average of the same quality with respect to the gold standard. This finding shows that the decrease of alignment accuracy is not necessarily a price for the computational efficiency.     

Functional proteome analysis of the banana plant (Musa spp.) using de novo sequence analysis of derivatized peptides

We report the use of chemical derivatization with MALDI-MS/MS analysis for de novo sequence analysis. Using three frequently used homology-based search algorithms, we were able to identify more than 40 proteins from banana, a nonmodel plant with unsequenced genome. Furthermore, this approach allowed the identification of different isoforms. We also observed that the identification score obtained varied according to the position of the peptide sequences in the query using the MS-Blast algorithm.     

A method to recognize distant repeats in protein sequences

An automated algorithm is presented that delineates protein sequence fragments which display similarity. The method incorporates a selection of a number of local nonoverlapping sequence alignments with the highest similarity scores and a graphtheoretical approach to elucidate the consistent start and end points of the fragments comprising one or more ensembles of related subsequences. The procedure allows the simultaneous identification of different types of repeats within one sequence. A multiple alignment of the resulting fragments is performed and a consensus sequence derived from the ensemble(s). Finally, a profile is constructed form the multiple alignment to detect possible and more distant members within the sequence. The method tolerates mutations in the repeats as well as insertions and deletions. The sequence spans between the various repeats or repeat clusters may be of different lengths. The technique has been applied to a number of proteins where the repeating fragments have been derived from information additional to the protein sequences. © 1993 Wiley-Liss, Inc.     

Extraction of symbolic determinants common to a family of biological sequences

A set of sequences can be defined by their common subsequences, and the length of these is a measure of the overall resemblance of the set. Each subsequence corresponds to a succession of symbols embedded in every sequence, following the same order but not necessarily contiguous. Determining the longest common subsequence (LCS) requires the exhaustive testing of all possible common subsequences, which sum up to about 2L, if L is the length of the shortest sequence. We present a polynomial algorithm (O(n X L4), where n is the number of sequences) for generating strings related to the LCS and constructed with the sequence alphabet and an indetermination symbol. Such strings are iteratively improved by deleting indetermination symbols and concomitantly introducing the greatest number of alphabet symbols. Processed accordingly, nucleic acid and protein sequences lead to key-words encompassing the salient positions of homologous chains, which can be used for aligning or classifying them, as well as for finding related sequences in data banks.     

Improving sequence-matching algorithms by working from both ends

Recent algorithms (e.g. Ukkonen, Fickett) align nucleic acid sequences (starting from the left) by bounding the allowed distance between subsequences by d, aligning, then incrementing d until all of both sequences are aligned. Aligning from both ends is more efficient. If the single-ended algorithm has computational cost CNk (C, k = constants; N = sequence length), the double-ended algorithm often has cost C(N/2)k.     

Analysis of the occurrence of promoter-sites in DNA.

We show that the occurrence and homology score (1) of promoter-sites in DNA depends upon the base composition of the DNA. We used simple probability theory to calculate the mean homology score expected for all promoter-sites that had a specific match in the canonical hexamers. By using the square root of this mean score as a measure of significance, we objectively classify all promoter-sites which are reported. We tested the theoretical approach in two ways. First, we used the program (PROMSEARCH) to analyze approximately 150,000 base pairs of random sequence DNA with different base compositions and we found excellent agreement with the theoretical predictions. Our second test was the analysis of a number of sequences drawn from the GENBANK DNA sequence database. We have analyzed 20 bacterial and bacteriophage sequences, which consisted of at least one operon, for promoter-sites. We found no absolute preference for promoter-sites within noncoding regions. We show the results of analyzing the phages lambda, T7 and fd, and the E. coli lac operon. The major known promoters in these sequences were all found correctly. We discuss the question of the location of a number of minor promoter-sites and show how PROMSEARCH can be used to help identify the correct location of the promoter. This approach can be applied to the search for any DNA site and should allow greater objectivity when comparing DNA sequences for meaningful subsequences.     

Local multiple alignment by consensus matrix

A new algorithm for aligning several sequences based on thecalculation of a consensus matrix and the comparison of allthe sequences using this consensus matrix is described. Thisconsensus matrix contains the preference scores of each nucleotideøaminoacid and gaps in every position of the alignment. Two modificationsof the algorithm corresponding to the evolutionary and functionalmeanings of the alignment were developed. The first one solvesthe best-fitting problem without any penalty for end gaps andwith an internal gap penalty function independent on the gaplength. This algorithm should be used when comparing evolutionary-relatedproteins for identifying the most conservative residues. Theother modification of the algorithm finds the most similar segmentsin the given sequences. It can be used for finding those partsof the sequences that are responsible for the same biologicalJunction. In this case the gap penalty function was chosen tobe proportional to the gap length. The result of aligning aminoacid sequences of neutral proteases and a compilation of 65allosteric effectors and substrates of PEP carboxylase are presented.     

Biological sequence analysis through the one-dimensional percolation transform and its enhanced version

MOTIVATION: The necessity to characterize the spatial uniformity (or lack of it) of symbols in biological sequences, given its implications for identification of the properties of the structures associated with the sequences. METHODS: A one-dimensional version of a recently introduced percolation-based approach is presented, which allows the accurate quantification of symbol distributions even in the presence of co-existing densities. An enhanced version of this methodology, which uses an agglomerative process to organize hierarchically the sequence into subsequences, is also proposed and illustrated. 3. RESULTS: The potential of the proposed methodology is illustrated with respect to synthetic and real data (1881 zebrafish and 1200 Xenopus proteins) and compared to two alternative multiscale methodologies, with encouraging results including the possibility to identify particularly remarkable amino acid arrangements in proteins. 4. CONTACT: luciano@if.sc.usp.br.