Journal of Bioinformatics and Computational Biology

This paper presents a composite multi-layer classifier system for predicting the subcellular localization of proteins based on their amino acid sequence. The work is an extension of our previous predictor PProwler v1.1 which is itself built upon the series of predictors SignalP and TargetP. In this study we outline experiments conducted to improve the classifier design. The major improvement came from using Support Vector machines as a "smart gate" sorting the outputs of several different targeting peptide detection networks. Our final model (PProwler v1.2) gives MCC values of 0.873 for non-plant and 0.849 for plant proteins. The model improves upon the accuracy of our previous subcellular localization predictor (PProwler v1.1) by 2% for plant data (which represents 7.5% improvement upon TargetP).

With the advent of experimental technologies like chemical cross-linking, it has become possible to obtain distances between specific residues of a newly sequenced protein. These types of experiments usually are less time consuming than X-ray crystallography or NMR. Consequently, it is highly desired to develop a method that incorporates this distance information to improve the performance of protein threading methods. However, protein threading with profiles in which constraints on distances between residues are given is known to be NP-hard. By using the notion of a maximum edge-weight clique finding algorithm, we introduce a more efficient method called FTHREAD for profile threading with distance constraints that is 18 times faster than its predecessor CLIQUETHREAD. Moreover, we also present a novel practical algorithm NTHREAD for profile threading with Non-strict constraints. The overall performance of FTHREAD on a data set shows that although our algorithm uses a simple threading function, our algorithm performs equally well as some of the existing methods. Particularly, when there are some unsatisfied constraints, NTHREAD (Non-strict constraints threading algorithm) performs better than threading with FTHREAD (Strict constraints threading algorithm). We have also analyzed the effects of using a number of distance constraints. This algorithm helps the enhancement of alignment quality between the query sequence and template structure, once the corresponding template structure is determined for the target sequence.

Pevzner and Sze¹⁹ have introduced the Planted (l,d)-Motif Problem to find similar patterns (motifs) in sequences which represent the promoter regions of co-regulated genes, where l is the length of the motif and d is the maximum Hamming distance around the similar patterns. Many algorithms have been developed to solve this motif problem. However, these algorithms either have long running times or do not guarantee the motif can be found. In this paper, we introduce new algorithms to solve this motif problem. Our algorithms can find motifs in reasonable time for not only the challenging (9, 2), (11, 3), (15, 5)-motif problems but for even longer motifs, say (20, 7), (30, 11) and (40, 15), which have never been seriously attempted by other researchers because of the large time and space required. Besides, our algorithms can be extended to find more complicated motifs structure called cis-regulatory modules (CRM).

To construct a phylogenetic tree or phylogenetic network for describing the evolutionary history of a set of species is a well-studied problem in computational biology. One previously proposed method to infer a phylogenetic tree/network for a large set of species is by merging a collection of known smaller phylogenetic trees on overlapping sets of species so that no (or as little as possible) branching information is lost. However, little work has been done so far on inferring a phylogenetic tree/network from a specified set of trees when in addition, certain evolutionary relationships among the species are known to be highly unlikely. In this paper, we consider the problem of constructing a phylogenetic tree/network which is consistent with all of the rooted triplets in a given set and none of the rooted triplets in another given set . Although NP-hard in the general case, we provide some efficient exact and approximation algorithms for a number of biologically meaningful variants of the problem.

Human ribonuclease A (RNaseA) superfamily consists of eight RNases with high similarity in which RNase2 and RNase3 share 76.7% identity. The evolutionary variation of RNases results in differential structures and functions of the enzymes. To distinguish the characteristics of each RNase, we developed reinforced merging algorithms (RMA) to rapidly identify the unique peptide motifs for each member of the highly conserved human RNaseA superfamily. Many motifs in RNase3 identified by RMA correlated well with the antigenic regions predicted by DNAStar. Two unique peptide motifs were experimentally confirmed to contain epitopes for monoclonal antibodies (mAbs) specifically against RNase3. Further analysis of homologous RNases in different species revealed that the unique peptide motifs were located at the correspondent positions, and one of these motifs indeed matched the epitope for a specific anti-bovine pancreatic RNaseA (bpRNaseA) antibody. Our method provides a useful tool for identification of unique peptide motifs for further experimental design. The RMA system is available and free for academic use at and .

New peptide encoding schemes are proposed to use with support vector machines for the direct recognition of T cell epitopes. The methods enable the presentation of information on (1) amino acid positions in peptides, (2) neighboring side chain interactions, and (3) the similarity between amino acids through a BLOSUM matrix. A procedure of feature selection is also introduced to strengthen the prediction. The computational results demonstrate competitive performance over previous techniques.

Transmembrane proteins affect vital cellular functions and pathogenesis, and are a focus of drug design. It is difficult to obtain diffraction quality crystals to study transmembrane protein structure. Computational tools for transmembrane protein topology prediction fill in the gap between the abundance of transmembrane proteins and the scarcity of known membrane protein structures. Their prediction accuracy is still inadequate: TMHMM, the current state-of-the-art method, has less than 52% accuracy in topology prediction on one set of transmembrane proteins of known topology. Based on the observation that there are functional domains that occur preferentially internal or external to the membrane, we have extended the model of TMHMM to incorporate functional domains, using a probabilistic approach originally developed for computational gene finding. Our extension is better than TMHMM in predicting the topology of transmembrane proteins. As prediction of functional domain improves, our system's prediction accuracy will likely improve as well.

In proteomics, tandem mass spectrometry is the key technology for peptide sequencing. However, partially due to the deficiency of peptide identification software, a large portion of the tandem mass spectra are discarded in almost all proteomics centers because they are not interpretable. The problem is more acute with the lower quality data from low end but more popular devices such as the ion trap instruments.

In order to deal with the noisy and low quality data, this paper develops a systematic machine learning approach to construct a robust linear scoring function, whose coefficients are determined by a linear programming. A prototype, PRIMA, was implemented. When tested with large benchmarks of varying qualities, PRIMA consistently has higher accuracy than commonly used software MASCOT, SEQUEST and X! Tandem.

Knowledge of molecular biological systems is increasing at an amazing pace. It is becoming harder to intuitively evaluate the significance of each interaction between the molecules of the complex biological systems. Hence, we need to develop an efficient computational method to explore the biological mechanisms. In this study, we employed a hybrid functional Petri net in order to analyze mammalian circadian genetic control mechanisms, which consists of feedback loops of clock genes and generates endogenous near 24 h rhythms in mammals. We constructed a computational model based on the available biological data, and by using Genomic Object Net, we performed computer simulations of the time courses of clock gene transcription and translation. Although the original model successfully reproduced most of the circadian genetic control mechanisms, two discrepancies remained despite a wide selection of the parameters. We found that addition of a hypothetical path into the original model result in successful simulation of time courses and phase relationships among clock genes. This also demonstrates the usefulness of the hybrid functional Petri net approach to biological systems.

Redox-active disulfides are capable of being oxidized and reduced under physiological conditions. The enzymatic role of redox-active disulfides in thiol-disulfide reductases is well-known, but redox-active disulfides are also present in non-enzymatic protein structures where they may act as switches of protein function. Here, we examine disulfides linking adjacent β-strands (cross-strand disulfides), which have been reported to be redox-active. Our previous work has established that these cross-strand disulfides have high torsional energies, a quantity likely to be related to the ease with which the disulfide is reduced. We examine the relationship between conformations of disulfides and their location in protein secondary structures. By identifying the overlap between cross-strand disulfides and various conformations, we wish to address whether the high torsional energy of a cross-strand disulfide is sufficient to confer redox activity or whether other factors, such as the presence of the cross-strand disulfide in a strained β-sheet, are required.