An accurate prediction model of digenic interaction for estimating pathogenic gene pairs of human diseases.
Ontology highlight
ABSTRACT:
SUBMITTER: Yuan Y
PROVIDER: S-EPMC9289819 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundPredicting the effect of drug-drug interactions (DDIs) precisely is important for safer and more effective drug co-prescription. Many computational approaches to predict the effect of DDIs have been proposed, with the aim of reducing the effort of identifying these interactions in vivo or in vitro, but room remains for improvement in prediction performance.ResultsIn this study, we propose a novel deep learning model to predict the effect of DDIs more accurately.. The proposed model uses autoencoders and a deep feed-forward network that are trained using the structural similarity profiles (SSP), Gene Ontology (GO) term similarity profiles (GSP), and target gene similarity profiles (TSP) of known drug pairs to predict the pharmacological effects of DDIs. The results show that GSP and TSP increase the prediction accuracy when using SSP alone, and the autoencoder is more effective than PCA for reducing the dimensions of each profile. Our model showed better performance than the existing methods, and identified a number of novel DDIs that are supported by medical databases or existing research.ConclusionsWe present a novel deep learning model for more accurate prediction of DDIs and their effects, which may assist in future research to discover novel DDIs and their pharmacological effects.
| S-EPMC6685287 | biostudies-literature
Project description:Noonan syndrome (NS) is an autosomal-dominant disorder with variable expressivity and locus heterogeneity. Despite several RAS pathway genes were implicated in NS, 20-30% of patients remain without molecular diagnosis, suggesting the involvement of further genes or multiple mechanisms. Eight patients out of 60, negative for conventional NS mutation analysis, with heterogeneous NS phenotype were investigated by means of target resequencing of 26 RAS/MAPK pathway genes. A trio was further characterized by means of whole-exome sequencing. Protein modeling and in silico prediction of protein stability allowed to identify possible pathogenic RAS pathway variants in four NS patients. A new c.355T>C variant in LZTR1 was found in patient 43. Two patients co-inherited variants in LRP1 and LZTR1 (patient 53), or LRP1 and SOS1 genes (patient 67). The forth patient (56) carried a compound heterozygote of RASAL3 gene variants and also an A2ML1 variant. While these subclinical variants are singularly present in healthy parents, they co-segregate in patients, suggesting their addictive effect and supporting a digenic inheritance, as alternative model to a more common monogenic transmission. The ERK1/2 and SAPK/JNK activation state, assessed on immortalized lymphocytes from patients 53 and 67 showed highest phosphorylation levels compared to their asymptomatic parents. These findings together with the lack of their co-occurrence in the 1000Genomes database strengthen the hypothesis of digenic inheritance in a subset of NS patients. This study suggests caution in the exclusion of subclinical variants that might play a pathogenic role providing new insights for alternative hereditary mechanisms.
| S-EPMC7608271 | biostudies-literature
Project description:We present FATHMM-XF, a method for predicting pathogenic point mutations in the human genome. Drawing on an extensive feature set, FATHMM-XF outperforms competitors on benchmark tests, particularly in non-coding regions where the majority of pathogenic mutations are likely to be found.The FATHMM-XF web server is available at http://fathmm.biocompute.org.uk/fathmm-xf/, and as tracks on the Genome Tolerance Browser: http://gtb.biocompute.org.uk. Predictions are provided for human genome version GRCh37/hg19. The data used for this project can be downloaded from: http://fathmm.biocompute.org.uk/fathmm-xf/.mark.rogers@bristol.ac.uk or c.campbell@bristol.ac.uk.Supplementary data are available at Bioinformatics online.
| S-EPMC5860356 | biostudies-other
Project description:To further our understanding of the complexity and genetic heterogeneity of rare diseases, it has become essential to shed light on how combinations of variants in different genes are responsible for a disease phenotype. With the appearance of a resource on digenic diseases, it has become possible to evaluate how digenic combinations differ in terms of the phenotypes they produce. All instances in this resource were assigned to two classes of digenic effects, annotated as true digenic and composite classes. Whereas in the true digenic class variants in both genes are required for developing the disease, in the composite class, a variant in one gene is sufficient to produce the phenotype, but an additional variant in a second gene impacts the disease phenotype or alters the age of onset. We show that a combination of variant, gene and higher-level features can differentiate between these two classes with high accuracy. Moreover, we show via the analysis of three digenic disorders that a digenic effect decision profile, extracted from the predictive model, motivates why an instance was assigned to either of the two classes. Together, our results show that digenic disease data generates novel insights, providing a glimpse into the oligogenic realm.
| S-EPMC5587785 | biostudies-literature
Project description:Some genetic diseases ("digenic traits") are due to the interaction between two DNA variants, which presumably reflects biochemical interactions. For example, certain forms of Retinitis Pigmentosa, a type of blindness, occur in the presence of two mutant variants, one each in the ROM1 and RDS genes, while the occurrence of only one such variant results in a normal phenotype. Detecting variant pairs underlying digenic traits by standard genetic methods is difficult and is downright impossible when individual variants alone have minimal effects. Frequent pattern mining (FPM) methods are known to detect patterns of items. We make use of FPM approaches to find pairs of genotypes (from different variants) that can discriminate between cases and controls. Our method is based on genotype patterns of length two, and permutation testing allows assigning p-values to genotype patterns, where the null hypothesis refers to equal pattern frequencies in cases and controls. We compare different interaction search approaches and their properties on the basis of published datasets. Our implementation of FPM to case-control studies is freely available.
| S-EPMC8391494 | biostudies-literature
Project description:There is an increasing need to use genome and transcriptome sequencing to genetically diagnose patients suffering from suspected monogenic rare diseases. The proper detection of compound heterozygous variant combinations as disease-causing candidates is a challenge in diagnostic workflows as haplotype information is lost by currently used next-generation sequencing technologies. Consequently, computational tools are required to phase, or resolve the haplotype of, the high number of heterozygous variants in the exome or genome of each patient. Here we present SmartPhase, a phasing tool designed to efficiently reduce the set of potential compound heterozygous variant pairs in genetic diagnoses pipelines. The phasing algorithm of SmartPhase creates haplotypes using both parental genotype information and reads generated by DNA or RNA sequencing and is thus well suited to resolve the phase of rare variants. To inform the user about the reliability of a phasing prediction, it computes a confidence score which is essential to select error-free predictions. It incorporates existing haplotype information and applies logical rules to determine variants that can be excluded as causing a recessive, monogenic disease. SmartPhase can phase either all possible variant pairs in predefined genetic loci or preselected variant pairs of interest, thus keeping the focus on clinically relevant results. We compared SmartPhase to WhatsHap, one of the leading comparable phasing tools, using simulated data and a real clinical cohort of 921 patients. On both data sets, SmartPhase generated error-free predictions using our derived confidence score threshold. It outperformed WhatsHap with regard to the percentage of resolved pairs when parental genotype information is available. On the cohort data, SmartPhase enabled on average the exclusion of approximately 22% of the input variant pairs in each singleton patient and 44% in each trio patient. SmartPhase is implemented as an open-source Java tool and freely available at http://ibis.helmholtz-muenchen.de/smartphase/.
| S-EPMC7032729 | biostudies-literature
Project description:Leukemia is a leading cause of cancer deaths in the developed countries. Great efforts have been undertaken in search of diagnostic biomarkers of leukemia. However, leukemia is highly complex and heterogeneous, involving interaction among multiple molecular components. Individual molecules are not necessarily sensitive diagnostic indicators. Network biomarkers are considered to outperform individual molecules in disease characterization. We applied an integrative approach that identifies active network modules as putative biomarkers for leukemia diagnosis. We first reconstructed the leukemia-specific PPI network using protein-protein interactions from the Protein Interaction Network Analysis (PINA) and protein annotations from GeneGo. The network was further integrated with gene expression profiles to identify active modules with leukemia relevance. Finally, the candidate network-based biomarker was evaluated for the diagnosing performance. A network of 97 genes and 400 interactions was identified for accurate diagnosis of leukemia. Functional enrichment analysis revealed that the network biomarkers were enriched in pathways in cancer. The network biomarkers could discriminate leukemia samples from the normal controls more effectively than the known biomarkers. The network biomarkers provide a useful tool to diagnose leukemia and also aids in further understanding the molecular basis of leukemia.
| S-EPMC5327377 | biostudies-literature
Project description:MotivationComputationally predicting major histocompatibility complex (MHC)-peptide binding affinity is an important problem in immunological bioinformatics. Recent cutting-edge deep learning-based methods for this problem are unable to achieve satisfactory performance for MHC class II molecules. This is because such methods generate the input by simply concatenating the two given sequences: (the estimated binding core of) a peptide and (the pseudo sequence of) an MHC class II molecule, ignoring biological knowledge behind the interactions of the two molecules. We thus propose a binding core-aware deep learning-based model, DeepMHCII, with a binding interaction convolution layer, which allows to integrate all potential binding cores (in a given peptide) with the MHC pseudo (binding) sequence, through modeling the interaction with multiple convolutional kernels.ResultsExtensive empirical experiments with four large-scale datasets demonstrate that DeepMHCII significantly outperformed four state-of-the-art methods under numerous settings, such as 5-fold cross-validation, leave one molecule out, validation with independent testing sets and binding core prediction. All these results and visualization of the predicted binding cores indicate the effectiveness of our model, DeepMHCII, and the importance of properly modeling biological facts in deep learning for high predictive performance and efficient knowledge discovery.Availability and implementationDeepMHCII is publicly available at https://github.com/yourh/DeepMHCII.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC9235502 | biostudies-literature
Project description:Intrinsically disordered proteins and regions (IDPs/IDRs) are functionally important proteins and regions that exist as highly dynamic conformations under natural physiological conditions. IDPs/IDRs exhibit a broad range of molecular functions, and their functions involve binding interactions with partners and remaining native structural flexibility. The rapid increase in the number of proteins in sequence databases and the diversity of disordered functions challenge existing computational methods for predicting protein intrinsic disorder and disordered functions. A disordered region interacts with different partners to perform multiple functions, and these disordered functions exhibit different dependencies and correlations. In this study, we introduce DisoFLAG, a computational method that leverages a graph-based interaction protein language model (GiPLM) for jointly predicting disorder and its multiple potential functions. GiPLM integrates protein semantic information based on pre-trained protein language models into graph-based interaction units to enhance the correlation of the semantic representation of multiple disordered functions. The DisoFLAG predictor takes amino acid sequences as the only inputs and provides predictions of intrinsic disorder and six disordered functions for proteins, including protein-binding, DNA-binding, RNA-binding, ion-binding, lipid-binding, and flexible linker. We evaluated the predictive performance of DisoFLAG following the Critical Assessment of protein Intrinsic Disorder (CAID) experiments, and the results demonstrated that DisoFLAG offers accurate and comprehensive predictions of disordered functions, extending the current coverage of computationally predicted disordered function categories. The standalone package and web server of DisoFLAG have been established to provide accurate prediction tools for intrinsic disorders and their associated functions.
| S-EPMC10762911 | biostudies-literature
Project description:The etiology of hypertriglyceridemia (HTG) and acute pancreatitis (AP) is complex. Herein, we dissected the underlying etiology in a patient with HTG and AP. The patient had a 20-year history of heavy alcohol consumption and an 8-year history of mild HTG. He was hospitalized for alcohol-triggered AP, with a plasma triglyceride (TG) level up to 21.4 mmol/L. A temporary rise in post-heparin LPL concentration (1.5-2.5 times of controls) was noted during the early days of AP whilst LPL activity was consistently low (50∼70% of controls). His TG level rapidly decreased to normal in response to treatment, and remained normal to borderline high during a ∼3-year follow-up period during which he had abstained completely from alcohol. Sequencing of the five primary HTG genes (i.e., LPL, APOC2, APOA5, GPIHBP1 and LMF1) identified two heterozygous variants. One was the common APOA5 c.553G > T (p.Gly185Cys) variant, which has been previously associated with altered TG levels as well as HTG-induced acute pancreatitis (HTG-AP). The other was a rare variant in the LPL gene, c.756T > G (p.Ile252Met), which was predicted to be likely pathogenic and found experimentally to cause a 40% loss of LPL activity without affecting either protein synthesis or secretion. We provide evidence that both a gene-gene interaction (between the common APOA5 variant and the rare LPL variant) and a gene-environment interaction (between alcohol and digenic inheritance) might have contributed to the development of mild HTG and alcohol-triggered AP in the patient, thereby improving our understanding of the complex etiology of HTG and HTG-AP.
| S-EPMC8143378 | biostudies-literature