Abstract
Motivation
Structure based ligand discovery is one of the most successful approaches for augmenting the drug discovery process. Currently, there is a notable shift towards machine learning ...(ML) methodologies to aid such procedures. Deep learning has recently gained considerable attention as it allows the model to 'learn' to extract features that are relevant for the task at hand.
Results
We have developed a novel deep neural network estimating the binding affinity of ligand-receptor complexes. The complex is represented with a 3D grid, and the model utilizes a 3D convolution to produce a feature map of this representation, treating the atoms of both proteins and ligands in the same manner. Our network was tested on the CASF-2013 'scoring power' benchmark and Astex Diverse Set and outperformed classical scoring functions.
Availability and implementation
The model, together with usage instructions and examples, is available as a git repository at http://gitlab.com/cheminfIBB/pafnucy.
Supplementary information
Supplementary data are available at Bioinformatics online.
MicroRNAs (miRNAs) represent a class of small non-coding RNAs that act as efficient gene expression regulators and thus play many important roles in living organisms. Due to their involvement in ...several known human pathological and pathogenic states, miRNA molecules have become an important issue in medicine and gained the attention of scientists from the pharmaceutical industry. In recent few years, a growing number of studies have provided evidence that miRNAs may be transferred from one species to another and regulate gene expression in the recipients' cells. The most intriguing results revealed that stable miRNAs derived from food plants may enter the mammals' circulatory system and, after reaching the target, inhibit the production of specific mammalian protein. Part of the scientific community has perceived this as an attractive hypothesis that may provide a foundation for novel therapeutic approaches. In turn, others are convinced about the "false positive" effect of performed experiments from which the mentioned results were achieved. In this article, we review the recent literature that provides evidence (from both fronts) of dietary, plant miRNA uptake and functionality in various consumers. Additionally, we discuss possible miRNA transport mechanisms from plant food sources to human cells.
MicroRNAs (miRNAs) are a class of small RNA molecules that regulate gene expression by inhibiting the protein translation or targeting the mRNA cleavage. They play many important roles in living ...organism cells; however, the knowledge on miRNAs functions has become more extensive upon their identification in biological fluids and recent reports on plant-origin miRNAs abundance in human plasma and serum. Considering these findings, we performed a rigorous bioinformatics analysis of publicly available, raw data from high-throughput sequencing studies on miRNAs composition in human and porcine breast milk exosomes to identify the fraction of food-derived miRNAs. Several processing and filtering steps were applied to increase the accuracy, and to avoid false positives. Through aforementioned analysis, 35 and 17 miRNA species, belonging to 25 and 11 MIR families, were identified, respectively. In the human samples the highest abundance levels yielded the ath-miR166a, pab-miR951, ptc-miR472a and bdi-miR168, while in the porcine breast milk exosomes, the zma-miR168a, zma-miR156a and ath-miR166a have been identified in the largest amounts. The consensus prediction and annotation of potential human targets for select plant miRNAs suggest that the aforementioned molecules may interact with mRNAs coding several transcription factors, protein receptors, transporters and immune-related proteins, thus potentially influencing human organism. Taken together, the presented analysis shows proof of abundant plant miRNAs in mammal breast milk exosomes, pointing at the same time to the new possibilities arising from this discovery.
Breast milk is a natural food and important component of infant nutrition. Apart from the alimentary substances, breast milk contains many important bioactive compounds, including endogenous microRNA ...molecules (miRNAs). These regulatory molecules were identified in various mammalian biological fluids and were shown to be mostly packed in exosomes. Recently, it was revealed that plant food-derived miRNAs are stably present in human blood and regulate the expression of specific human genes. Since then, the scientific community has focused its efforts on contradicting or confirming this discovery. With the same intention, qRT-PCR experiments were performed to evaluate the presence of five plant food-derived miRNAs (miR166a, miR156a, miR157a, miR172a and miR168a) in breast milk (whole milk and exosomes) from healthy volunteers. In whole milk samples, all examined miRNAs were identified, while only two of these miRNAs were confirmed to be present in exosomes. The plant miRNA concentration in the samples ranged from 4 to 700 fM. Complementary bioinformatics analysis suggests that the evaluated plant miRNAs may potentially influence several crucial biological pathways in the infant organism.
In recent years machine learning (ML) took bio- and cheminformatics fields by storm, providing new solutions for a vast repertoire of problems related to protein sequence, structure, and interactions ...analysis. ML techniques, deep neural networks especially, were proven more effective than classical models for tasks like predicting binding affinity for molecular complex. In this work we investigated the earlier stage of drug discovery process - finding druggable pockets on protein surface, that can be later used to design active molecules. For this purpose we developed a 3D fully convolutional neural network capable of binding site segmentation. Our solution has high prediction accuracy and provides intuitive representations of the results, which makes it easy to incorporate into drug discovery projects. The model's source code, together with scripts for most common use-cases is freely available at http://gitlab.com/cheminfIBB/kalasanty.
Peptides are fragments of proteins that carry out biological functions. They act as signaling entities via all domains of life and interfere with protein-protein interactions, which are indispensable ...in bio-processes. Short peptides include fundamental molecular information for a prelude to the symphony of life. They have aroused considerable interest due to their unique features and great promise in innovative bio-therapies. This work focusing on the current state-of-the-art short peptide-based therapeutical developments is the first global review written by researchers from all continents, as a celebration of 100 years of peptide therapeutics since the commencement of insulin therapy in the 1920s. Peptide "drugs" initially played only the role of hormone analogs to balance disorders. Nowadays, they achieve numerous biomedical tasks, can cross membranes, or reach intracellular targets. The role of peptides in bio-processes can hardly be mimicked by other chemical substances. The article is divided into independent sections, which are related to either the progress in short peptide-based theranostics or the problems posing challenge to bio-medicine. In particular, the SWOT analysis of short peptides, their relevance in therapies of diverse diseases, improvements in (bio)synthesis platforms, advanced nano-supramolecular technologies, aptamers, altered peptide ligands and in silico methodologies to overcome peptide limitations, modern smart bio-functional materials, vaccines, and drug/gene-targeted delivery systems are discussed.
Although translation is the key step during gene expression, it remains poorly characterized at the level of individual genes. For this reason, we developed Transimulation - a web service measuring ...translational activity of genes in three model organisms: Escherichia coli, Saccharomyces cerevisiae and Homo sapiens. The calculations are based on our previous computational model of translation and experimental data sets. Transimulation quantifies mean translation initiation and elongation time (expressed in SI units), and the number of proteins produced per transcript. It also approximates the number of ribosomes that typically occupy a transcript during translation, and simulates their propagation. The simulation of ribosomes' movement is interactive and allows modifying the coding sequence on the fly. It also enables uploading any coding sequence and simulating its translation in one of three model organisms. In such a case, ribosomes propagate according to mean codon elongation times of the host organism, which may prove useful for heterologous expression. Transimulation was used to examine evolutionary conservation of translational parameters of orthologous genes. Transimulation may be accessed at http://nexus.ibb.waw.pl/Transimulation (requires Java version 1.7 or higher). Its manual and source code, distributed under the GPL-2.0 license, is freely available at the website.
Filiano and Kinney proposed a triple-risk model for the sudden infant death syndrome (SIDS) that involves the intersection of three risks: (1) a vulnerable infant, (2) a critical developmental period ...in homeostatic control, and (3) an exogenous stressor(s). The primary evidence for the role of a critical developmental period in SIDS etiology is the peak of cases around the third month of life. Independently, several studies pointed to correlation between gestational age and age at death in SIDS, but used that to assess the SIDS risk for preterm infants, ignoring further ramifications.
We did a detailed analysis of CDC data spanning over two decades (1983-2011). We focused not only on the correlation between two age variables (gestational and age at death), but also on the possibility of misdiagnosis. Also, we attempted to account for potential biases in the data induced by the ICD-9/ICD-190 transition or the "Back to Sleep" campaign.
The peak of deaths in the third month of life, that was the main argument for the role of the critical development period, wasn't unique to SIDS. However, we confirmed an almost linear and negative correlation between gestational age and the week of death due to SIDS. This pattern (slope of correlation < 0 and significance of correlation p < 0.05) is characteristic of SIDS among all diseases analyzed in the study.
We interpret the results as the evidence of the role of the critical development period in SIDS etiology. Possibly more attention in the future research should be put to theories that are based on homeostatic control.
Translation is still poorly characterised at the level of individual proteins and its role in regulation of gene expression has been constantly underestimated. To better understand the process of ...protein synthesis we developed a comprehensive and quantitative model of translation, characterising protein synthesis separately for individual genes. The main advantage of the model is that basing it on only a few datasets and general assumptions allows the calculation of many important translational parameters, which are extremely difficult to measure experimentally. In the model, each gene is attributed with a set of translational parameters, namely the absolute number of transcripts, ribosome density, mean codon translation time, total transcript translation time, total time required for translation initiation and elongation, translation initiation rate, mean mRNA lifetime, and absolute number of proteins produced by gene transcripts. Most parameters were calculated based on only one experimental dataset of genome-wide ribosome profiling. The model was implemented in Saccharomyces cerevisiae, and its results were compared with available data, yielding reasonably good correlations. The calculated coefficients were used to perform a global analysis of translation in yeast, revealing some interesting aspects of the process. We have shown that two commonly used measures of translation efficiency - ribosome density and number of protein molecules produced - are affected by two distinct factors. High values of both measures are caused, i.a., by very short times of translation initiation, however, the origins of initiation time reduction are completely different in both cases. The model is universal and can be applied to any organism, if the necessary input data are available. The model allows us to better integrate transcriptomic and proteomic data. A few other possibilities of the model utilisation are discussed concerning the example of the yeast system.
PDF
