Human Protein Reference Database (HPRD) () was developed to serve as a comprehensive collection of protein features, post-translational modifications (PTMs) and protein–protein interactions. Since the original report, this database has increased to >20 000 proteins entries and has become the largest database for literature-derived protein–protein interactions (>30 000) and PTMs (>8000) for human proteins. We have also introduced several new features in HPRD including: (i) protein isoforms, (ii) enhanced search options, (iii) linking of pathway annotations and (iv) integration of a novel browser, GenProt Viewer (), developed by us that allows integration of genomic and proteomic information. With the continued support and active participation by the biomedical community, we expect HPRD to become a unique source of curated information for the human proteome and spur biomedical discoveries based on integration of genomic, transcriptomic and proteomic data.

Plasma is one of the best studied compartments in the human body and serves as an ideal body fluid for the diagnosis of diseases. This report provides a detailed functional annotation of all the plasma proteins identified to date. In all, gene products encoded by 3778 distinct genes were annotated based on proteins previously published in the literature as plasma proteins and the identification of multiple peptides from proteins under HUPO's Plasma Proteome Project. Our analysis revealed that 51% of these genes encoded more than one protein isoform. All single nucleotide polymorphisms involving protein-coding regions were mapped onto the protein sequences. We found a number of examples of isoform-specific subcellular localization as well as tissue expression. This database is an attempt at comprehensive annotation of a complex subproteome and is available on the web at http://www.plasmaproteomedatabase.org.

A quantum-classical molecular dynamics model (QCMD) designed for simulations of proton or electron transfer processes in molecular systems is described and applied to several model problems. The primary goal of this work is the elucidation of enzymatic reactions. For example, using the QCMD model, the dynamics of key protons in an enzyme's active site might be described by the time-dependent Schroedinger equation while the dynamics of the remaining atoms are described using MD. The coupling between the quantum proton(s) and the classical atoms is accomplished via extended Hellmann−Feynman forces, as well as the time dependence of the potential energy function in the Schroedinger equation. The potential energy function is either parametrized prior to the simulations or can be computed using a parametrized valence bond (VB) method (QCMD/VB model). The QCMD method was used to simulate proton transfer in a proton bound ammonia−ammonia dimer as well as to simulate dissociation of a Xe−HI complex in its electronic excited state. The simulation results are compared with data obtained using a quantum-classical time-dependent self-consistent field method (Q/C TDSCF) and with results of fully quantum-dynamical simulations. Finally QCMD/VB simulations of a hydrolytic process catalyzed by phospholipase A2, including quantum-dynamical dissociation of a water molecule in the active site, are reported. To the best of our knowledge, these are the first simulations that explicitly use the time-dependent Schroedinger equation to describe enzyme catalytic activity.

mAn approximate valence bond (AVB) method was parametrized at a microscopic level for proton transfer and hydroxyanion nucleophilic reactions in enzyme catalytic processes.The method was applied to describe hydrolytic activity of phospholipase A,. The AVB parametrization is based on density functional and conventional ab initio calculations calibrated with respect to experimental data in the gas phase. The method was used as a fast generator of the potential energy function in a quantum-classical molecular dynamics (QCMD) simulations describing atomic motions as well as propagation of the proton wave function in the enzyme active site. The protein environment surrounding the active site and solvent effects are included in the model via electrostatic interactions perturbing the original AVB Hamiltonian.

Self-consistent charge-density functional tight-binding SCC-DFTB is a computationally efficient method applicable to large (bio)molecular systems in which (bio)chemical reactions may occur. Among these reactions are proton transfer processes. This method, along with more advanced ab initio techniques, is applied in this study to compute intramolecular barriers for single and double proton transfer processes in the model systems, malonaldehyde and porphycene, respectively. SCC-DFTB is compared with experimental data and higher-level ab initio calculations. For malonaldehyde, the SCC-DFTB barrier height is 3.1 kcal/mol in vacuo and 4.2 kcal/mol in water solution. In the case of porphycene, the minimum energy pathways for double intramolecular proton transfer were determined using the conjugate peak refinement (CPR) method. Six isomers of porphycene were ordered according to energy. The only energetically allowed pathway was found to connect two symmetrical trans states via an unstable cis-A isomer. The SCC-DFTB barrier heights are 11.1 kcal/mol for the trans-cis-A

BackgroundDetection of highly divergent or yet unknown viruses from metagenomics sequencing datasets is a major bioinformatics challenge. When human samples are sequenced, a large proportion of assembled contigs are classified as “unknown”, as conventional methods find no similarity to known sequences. We wished to explore whether machine learning algorithms using Relative Synonymous Codon Usage frequency (RSCU) could improve the detection of viral sequences in metagenomic sequencing data.ResultsWe trained Random Forest and Artificial Neural Network using metagenomic sequences taxonomically classified into virus and non-virus classes. The algorithms achieved accuracies well beyond chance level, with area under ROC curve 0.79. Two codons (TCG and CGC) were found to have a particularly strong discriminative capacity.ConclusionRSCU-based machine learning techniques applied to metagenomic sequencing data can help identify a large number of putative viral sequences and provide an addition to conventional methods for taxonomic classification.Electronic supplementary materialThe online version of this article (10.1186/s12859-018-2340-x) contains supplementary material, which is available to authorized users.

No abstract

We present a detailed comparison of the efficiency and accuracy of the second-and third-order split operator methods, a time dependent modified Cayley method, and the Chebychev polynomial expansion method for solving the time dependent Schrodinger equation in the onedimensional double well potential energy function. We also examine the efficiency and accuracy of the split operator and modified Cayley methods for the imaginary time propagation.