Search

A number of machine learning (ML)-based algorithms have been proposed for predicting mutation-induced stability changes in proteins. In this critical review, we used hypothetical reverse mutations to evaluate the performance of five representative algorithms and found all of them suffer from the problem of overfitting. This approach is based on the fact that if a wild-type protein is more stable than a mutant protein, then the same mutant is less stable than the wild-type protein. We analyzed the underlying issues and suggest that the main causes of the overfitting problem include that the numbers of training cases were too small, and the features used in the models were not sufficiently informative for the task. We make recommendations on how to avoid overfitting in this important research area and improve the reliability and robustness of ML-based algorithms in general.

Motivation: Targeting peptides are the most important signal controlling the import of nuclear encoded proteins into mitochondria and plastids. In the lack of experimental information, their prediction is an essential step when proteomes are annotated for inferring both the localization and the sequence of mature proteins. Results: We developed TPpred a new predictor of organelle-targeting peptides based on Grammatical-Restrained Hidden Conditional Random Fields. TPpred is trained on a non-redundant dataset of proteins where the presence of a target peptide was experimentally validated, comprising 297 sequences. When tested on the 297 positive and some other 8010 negative examples, TPpred outperformed available methods in both accuracy and Matthews correlation index (96% and 0.58, respectively). Given its very low–false-positive rate (3.0%), TPpred is, therefore, well suited for large-scale analyses at the proteome level. We predicted that from ∼4 to 9% of the sequences of human, Arabidopsis thaliana and yeast proteomes contain targeting peptides and are, therefore, likely to be localized in mitochondria and plastids. TPpred predictions correlate to a good extent with the experimental annotation of the subcellular localization, when available. TPpred was also trained and tested to predict the cleavage site of the organelle-targeting peptide: on this task, the average error of TPpred on mitochondrial and plastidic proteins is 7 and 15 residues, respectively. This value is lower than the error reported by other methods currently available. Availability: The TPpred datasets are available at http://biocomp.unibo.it/∼valentina/TPpred/. TPpred is available on request from the authors. Contact: gigi@biocomp.unibo.it Supplementary information: Supplementary data are available at Bioinformatics online.

We introduce BAR-PLUS (BAR(+)), a web server for functional and structural annotation of protein sequences. BAR(+) is based on a large-scale genome cross comparison and a non-hierarchical clustering procedure characterized by a metric that ensures a reliable transfer of features within clusters. In this version, the method takes advantage of a large-scale pairwise sequence comparison of 13,495,736 protein chains also including 988 complete proteomes. Available sequence annotation is derived from UniProtKB, GO, Pfam and PDB. When PDB templates are present within a cluster (with or without their SCOP classification), profile Hidden Markov Models (HMMs) are computed on the basis of sequence to structure alignment and are cluster-associated (Cluster-HMM). Therefrom, a library of 10,858 HMMs is made available for aligning even distantly related sequences for structural modelling. The server also provides pairwise query sequence-structural target alignments computed from the correspondent Cluster-HMM. BAR(+) in its present version allows three main categories of annotation: PDB [with or without SCOP (*)] and GO and/or Pfam; PDB (*) without GO and/or Pfam; GO and/or Pfam without PDB (*) and no annotation. Each category can further comprise clusters where GO and Pfam functional annotations are or are not statistically significant. BAR(+) is available at http://bar.biocomp.unibo.it/bar2.0.

Approximately 10%-15% of gastrointestinal stromal tumors (GISTs) in adults do not harbor any mutation in the KIT or PDGFRA genes (ie, KIT/PDGFRA wild-type GISTs). Recently, mutations in SDHB and SDHC (which encode succinate dehydrogenase subunits B and C, respectively) but not in SDHA and SDHD (which encode subunits A and D, respectively) were identified in KIT/PDGFRA wild-type GISTs. To search for novel pathogenic mutations, we sequenced the tumor transcriptome of two young adult patients who developed sporadic KIT/PDGFRA wild-type GISTs by using a massively parallel sequencing approach. The only variants identified as disease related by computational analysis were in SDHA. One patient carried the homozygous nonsense mutation p.Ser384X, the other patient was a compound heterozygote harboring a p.Arg31X nonsense mutation and a p.Arg589Trp missense mutation. The heterozygous nonsense mutations in both patients were present in germline DNA isolated from peripheral blood. Protein structure analysis indicates that all three mutations lead to functional inactivation of the protein. This is the first report, to our knowle dge, that identifies SDHA inactivation as a common oncogenic event in GISTs that lack a mutation in KIT and PDGFRA.

Motivation: Searching for structural similarity is a key issue of protein functional annotation. The maximum contact map overlap (CMO) is one of the possible measures of protein structure similarity. Exact and approximate methods known to optimize the CMO are computationally expensive and this hampers their applicability to large-scale comparison of protein structures. Results: In this article, we describe a heuristic algorithm (Al-Eigen) for finding a solution to the CMO problem. Our approach relies on the approximation of contact maps by eigendecomposition. We obtain good overlaps of two contact maps by computing the optimal global alignment of few principal eigenvectors. Our algorithm is simple, fast and its running time is independent of the amount of contacts in the map. Experimental testing indicates that the algorithm is comparable to exact CMO methods in terms of the overlap quality, to structural alignment methods in terms of structure similarity detection and it is fast enough to be suited for large-scale comparison of protein structures. Furthermore, our preliminary tests indicates that it is quite robust to noise, which makes it suitable for structural similarity detection also for noisy and incomplete contact maps. Availability: Available at http://bioinformatics.cs.unibo.it/Al-Eigen Contact: dilena@cs.unibo.it Supplementary information: Supplementary data are available at Bioinformatics online.

Automated sequence annotation is a major goal of post-genomic era with hundreds of genomes in the databases, from both prokaryotes and eukaryotes. While the number of fully sequenced chromosomes from microbial organisms exponentially increased in the last decade above 600, presently we know the whole DNA content of only 25 eukaryotic organisms, including Homo sapiens. However, the process of genome annotation is far from being completed. This is particularly relevant in eukaryotes, whose cells contain several subcellular compartments, or organelles, enclosed by membranes, where different relevant functions are performed. Translocation across the membrane into the organelles is a highly regulated and complex cellular process. Indeed different proteins and/or protein isoforms, originated from genes by alternative splicing, may be conveyed to different cell compartments, depending on their specific role in the cell. During recent years the prediction of subcellular localization (SL) by computational means has been an active research area. Several methods are presently available based on different notions and addressing different aspects of SL. This review provides a short overview of the most well performing methods described in the literature, highlighting their predictive capabilities and different applications.

The detection of remote homolog pairs of proteins using computational methods is a pivotal problem in structural bioinformatics, aiming to compute protein folds on the basis of information in the database of known structures. In the last 25 years, several methods have been developed to tackle this problem, based on different approaches including sequence-sequence alignments and/or structure comparison. In this article, we will briefly discuss When, Why, Where and How (WWWH) to perform remote homology search, reviewing some of the most widely adopted computational approaches. The specific aim is highlighting the basic criteria implemented by different research groups and commenting on the status of the art as well as on still-open questions.

Motivation: Human single nucleotide polymorphisms (SNPs) are the most frequent type of genetic variation in human population. One of the most important goals of SNP projects is to understand which human genotype variations are related to Mendelian and complex diseases. Great interest is focused on non-synonymous coding SNPs (nsSNPs) that are responsible of protein single point mutation. nsSNPs can be neutral or disease associated. It is known that the mutation of only one residue in a protein sequence can be related to a number of pathological conditions of dramatic social impact such as Altzheimer's, Parkinson's and Creutzfeldt-Jakob's diseases. The quality and completeness of presently available SNPs databases allows the application of machine learning techniques to predict the insurgence of human diseases due to single point protein mutation starting from the protein sequence. Results: In this paper, we develop a method based on support vector machines (SVMs) that starting from the protein sequence information can predict whether a new phenotype derived from a nsSNP can be related to a genetic disease in humans. Using a dataset of 21 185 single point mutations, 61% of which are disease-related, out of 3587 proteins, we show that our predictor can reach more than 74% accuracy in the specific task of predicting whether a single point mutation can be disease related or not. Our method, although based on less information, outperforms other web-available predictors implementing different approaches. Availability: A beta version of the web tool is available at Contact: casadio@alma.unibo.it

TRAMPLE (http://gpcr.biocomp.unibo.it/biodec/) is a web application server dedicated to the detection and the annotation of transmembrane protein sequences. TRAMPLE includes different state-ofthe- art algorithms for the prediction of signal peptides, transmembrane segments (both beta-strands and alpha-helices), secondary structure and fast fold recognition. TRAMPLE also includes a complete content management system to manage the results of the predictions. Each user of the server has his/ her own workplace, where the data can be stored, organized, accessed and annotated with documents through a simple web-based interface. In this manner, TRAMPLE significantly improves usability with respect to other more traditional web servers.

Motivation: One important requirement for protein design is to be able to predict changes of protein stability upon mutation. Different methods addressing this task have been described and their performance tested considering global linear correlation between predicted and experimental data. Neither is direct statistical evaluation of their prediction performance available, nor is a direct comparison among different approaches possible. Recently, a significant database of thermodynamic data on protein stability changes upon single point mutation has been generated (ProTherm). This allows the application of machine learning techniques to predicting free energy stability changes upon mutation starting from the protein sequence. Results: In this paper, we present a neural-network-based method to predict if a given mutation increases or decreases the protein thermodynamic stability with respect to the native structure. Using a dataset consisting of 1615 mutations, our predictor correctly classifies >80% of the mutations in the database. On the same task and using the same data, our predictor performs better than other methods available on the Web. Moreover, when our system is coupled with energy-based methods, the joint prediction accuracy increases up to 90%, suggesting that it can be used to increase also the performance of pre-existing methods, and generally to improve protein design strategies. Availability: The server is under construction and will be available at http://www.biocomp.unibo.it

We describe the computation of a model of the thermophilic NAD-dependent homotetrameric alcohol dehydrogenase from the archaeon Sulfolobus solfataricus (SsADH). Modeling is based on the knowledge that each monomer contains two Zn ions with catalytic and structural function, respectively. In the database of known structures, proteins with similar functions are either dimers containing two zinc ions per monomer or tetramers with one zinc ion per monomer. In any case, the sequence identity of the target to the possible templates is low. A threading procedure is therefore developed which includes constraints taking into account residue conservation both at the zinc ion binding and at the monomer-monomer interaction sites in the tetrameric unit. The model is consistent with previously reported data. Furthermore, cross-linking experiments are described which support the computed tetrameric model.

Summary: Fault Tolerant Contact Map Reconstruction (FT-COMAR) is a heuristic algorithm for the reconstruction of the protein three-dimensional structure from (possibly) incomplete (i.e. containing unknown entries) and noisy contact maps. FT-COMAR runs within minutes, allowing its application to a large-scale number of predictions. Availability: http://bioinformatics.cs.unibo.it/FT-COMAR Contact: vassura@cs.unibo.it Supplementary information: Supplementary data are available on Bioinformatics online.