Computational prediction of disease related lncRNAs using machine learning.

Long non-coding RNAs (lncRNAs), which were once considered as transcriptional noise, are now in the limelight of current research. LncRNAs play a major role in regulating various biological processes such as imprinting, cell differentiation, and splicing. The mutations of lncRNAs are involved in various complex diseases. Identifying lncRNA-disease associations has gained a lot of attention as predicting it efficiently will lead towards better disease treatment. In this study, we have developed a machine learning model that predicts disease-related lncRNAs by combining sequence and structure-based features. The features were trained on SVM and Random Forest classifiers. We have compared our method with the state-of-the-art and obtained the highest F1 score of 76% on SVM classifier. Moreover, this study has overcome two serious limitations of the reported method which are lack of redundancy checking and implementation of oversampling for balancing the positive and negative class. Our method has achieved improved performance among machine learning models reported for lncRNA-disease associations. Combining multiple features together specifically lncRNAs sequence mutation has a significant contribution to the disease related lncRNA prediction.

Clinically significant findings of high-risk mutations in human SLC29A4 gene associated with diabetes mellitus type 2 in Pakistani population.

This study conducted an in-depth analysis combining computational and experimental verifications of the deleterious missense mutations associated with the SLC29A4 protein. The functional annotation of the non-synonymous single nucleotide polymorphism (nsSNPs), followed by structure-function analysis, revealed 13 single nucleotide polymorphisms (SNP) as the most damaging. Among these, six mutants P429T/S, L144S, M108V, N86H, and V79E, were predicted as structurally and functionally damaging by protein stability analysis. Also, these variants are located at evolutionary conserved regions, either buried, contributing to the structural damage, or exposed, causing functional changes in the protein. These mutants were further taken for molecular docking studies. When verified via experimental analysis, the SNPs M108V (rs149798710), N86H (rs151039853), and V79E (rs17854505) showed an association with type 2 diabetes mellitus (T2DM). Minor allele frequency for rs149798710 (A > G) was 0.23 in controls, 0.29 in metformin responders, 0.37 in metformin non-responder, for rs151039853 (A > C) was 0.21 in controls, 0.28 in metformin responders, 0.36 in metformin non-responder and for rs17854505 (T > A) was 0.20 in controls, 0.25 in metformin responders, 0.37 in metformin non-responder. Hence, this study concludes that SLC29A4 M108V (rs149798710), N86H (rs151039853), and V79E (rs17854505) polymorphisms were associated with the increased risk of T2DM as well as with the increased risk towards the failure of metformin therapeutic response in T2DM patients of Pakistan. Communicated by Ramaswamy H. Sarma.

Identification of destabilizing SNPs in SARS-CoV2-ACE2 protein and spike glycoprotein: implications for virus entry mechanisms.

COVID-19 an outbreak of a novel corona virus originating from Wuhan, China in December 2019 has now spread across the entire world and has been declared a pandemic by WHO. Angiotensin converting enzyme 2 (ACE2) is a receptor protein that interacts with the spike glycoprotein of the host to facilitate the entry of coronavirus (SARS-CoV-2) hence causing the disease (COVID-19). Our experimental design is based on bioinformatics approach that combines sequence, structure and consensus based tools to label a protein coding single nucleotide polymorphism (SNP) as damaging/deleterious or neutral. The interaction of wildtype ACE2-spike glycoprotein and their variants were analyzed using docking studies. The mutations W461R, G405E and F588S in ACE2 receptor protein and population specific mutations P391S, C12S and G1223A in the spike glycoprotein were predicted as highly destabilizing to the structure of the bound complex. So far, no extensive in silico study has been reported that identifies the effect of SNPs on Spike glycoprotein-ACE2 interaction exploring both sequence and structural features. To this end, this study conducted an in-depth analysis that facilitates in identifying the mutations that blocks the interaction of two proteins that can result in stopping the virus from entering the host cell.Communicated by Ramaswamy H. Sarma.

Computational Prediction of lncRNA-Protein Interactions using Machine learning.

Long non-coding RNAs have generated much scientific interest because of their functional significance in regulating various biological processes and also their dysfunction has been implicated in disease progression. LncRNAs usually bind with proteins to perform their function. The experimental approaches for identifying these interactions are time taking and expensive. Lately, numerous method on predicting lncRNA-protein interactions have been reported yet, they all have some prevalent drawbacks that limit their prediction performance. In this research, we proposed a computational method based on a similarity scheme that integrates features derived from sequence and structure similarities. When compared with the state of the art, the proposed method has achieved highest performance with accuracy and F1 measure of 98.6% and 98.7% using XGBoost as classifier. Our results showed that by combining sequence and structure based features the lncRNA protein interactions can be better predicted and can also complement the experimental techniques for this task.Clinical Relevance- The lncRNA-protein interactions play significant role in regulating various biological processes. This can help in providing early diagnosis and better treatment for cancer related diseases.

Evaluating the cleavage efficacy of CRISPR-Cas9 sgRNAs targeting ineffective regions of Arabidopsis thaliana genome.

The CRISPR-Cas9 system has recently evolved as a powerful mutagenic tool for targeted genome editing. The impeccable functioning of the system depends on the optimal design of single guide RNAs (sgRNAs) that mainly involves sgRNA specificity and on-target cleavage efficacy. Several research groups have designed algorithms and models, trained on mammalian genomes, for predicting sgRNAs cleavage efficacy. These models are also implemented in most plant sgRNA design tools due to the lack of on-target cleavage efficacy studies in plants. However, one of the major drawbacks is that almost all of these models are biased for considering only coding regions of the DNA while excluding ineffective regions, which are of immense importance in functional genomics studies especially for plants, thus making prediction less reliable. In the present study, we evaluate the on-target cleavage efficacy of experimentally validated sgRNAs designed against diverse ineffective regions of Arabidopsis thaliana genome using various statistical tests. We show that nucleotide preference in protospacer adjacent motif (PAM) proximal region, GC content in the PAM proximal seed region, intact RAR and 3rd stem loop structures, and free accessibility of nucleotides in seed and tracrRNA regions of sgRNAs are important determinants associated with their high on-target cleavage efficacy. Thus, our study describes the features important for plant sgRNAs high on-target cleavage efficacy against ineffective genomic regions previously shown to give rise to ineffective sgRNAs. Moreover, it suggests the need of developing an elaborative plant-specific sgRNA design model considering the entire genomic landscape including ineffective regions for enabling highly efficient genome editing without wasting time and experimental resources.

A comprehensive study on identifying the structural and functional SNPs of human neuronal membrane glycoprotein M6A (GPM6A).

Glycoprotein M6A, a stress related gene, plays an important role in synapse and filopodia formation. Filopodia formation is vital for development, immunity, angiogenesis, wound healing and metastasis. In this study, structural and functional analysis of high-risk SNPs associated with Glycoprotein M6-A were evaluated using six different bioinformatics tools. Results classified T210I, T134I, Y153H, I215T, F156L, T160I, I226T, R247W, R178C, W159R, N157S and P151L as deleterious mutants that are crucial for the structure and function of the protein causing malfunction of M6-a and ultimately leads to disease development. The three-dimensional structure of wild-type M6-a and mutant M6-a were also predicted. Furthermore, the effects of high risk substitutions were also analyzed with interaction with valproic acid. Based on structural models obtained, the binding pocket of ligand bound glycoprotein M6-A structure showed few core interacting residues which are different in the mutant models. Among all substitutions, F156L showed complete loss of binding pocket when interacting with valproic acid as compared to the wild type model. Up to the best of our knowledge this is the first comprehensive study where GPM6A mutations were analyzed. The mechanism of action of GPM6A is still not fully defined which limits the understanding of functional details encoding M6-A. Our results may help enlighten some molecular aspects underlying glycoprotein M6-A. Communicated by Ramaswamy H. Sarma.

Classification and Computational Analysis of Arabidopsis thaliana Sperm Cell-Specific F-Box Protein Gene 3p.AtFBP113.

In plants, F-box proteins (FBPs) constitute one of the largest superfamilies of regulatory proteins. Most F-box proteins are shown to be an integral part of SCF complexes, which carry out the degradation of proteins and regulate diverse important biological processes. Anthers and pollen development have a huge importance in crop breeding. Despite the vast diversity of FBPs in Arabidopsis male reproductive organs, their role in anther and pollen development is not much explored. Moreover, a standard nomenclature for naming FBPs is also lacking. Here, we propose a standard nomenclature for naming the FBPs of Arabidopsis thaliana uniformly and carry out a systematic analysis of sperm cell-specific FBP gene, i.e., 3p.AtFBP113 due to its reported high and preferential expression, for detailed functional annotation. The results revealed that 3p.AtFBP113 is located on the small arm of chromosome and encodes 397 amino acid long soluble, stable, and hydrophilic protein with the possibility of localization in various cellular compartments. The presence of the C-terminal F-box associated domain (FBA) with immunoglobulin-like fold anticipated its role in protein binding. Gene ontology based functional annotation and tissue-specific gene co-expression analysis further strengthened its role in protein binding and ubiquitination. Moreover, various potential post/co-translational modifications were anticipated and the predicted tertiary structure also showed the presence of characteristic domains and fold. Thus, the outcomes of the study will be useful in developing a better understating of the function of 3p.AtFBP113 during the process of pollen development, which will be helpful for targeting the gene for manipulation of male fertility that has immense importance in hybrid breeding.

Effects of Single-Nucleotide Polymorphisms in Calmodulin-Dependent Protein Kinase Kinase 2 (CAMKK2): A Comprehensive Study.

Calmodulin-dependent protein kinase kinase 2 (CAMKK2) is a protein kinase that belongs to the serine/threonine kinase family. It phosphorylates kinases like CAMK1, CAMK2, and AMP, and this signaling cascade is involved in various biological processes including cell proliferation, apoptosis, and proliferation. Also, the CAMKK2 signaling activity is required for the healthy activity of the brain which otherwise can cause diseases like bipolar disorders and anxiety. The current study is based on in silico bioinformatics analysis that combines sequence- and structure-based predictions to mark a SNP as damaging or neutral. The combined results from sequence-based, evolutionary conservation-based, and consensus-based tools have predicted a total of 18 nsSNPs as deleterious, and these nsSNPs were further subjected to structure-based analysis. The six mutant models of V195A, V249M, R311C, F366Y, P389T, and W445C showed a higher deviation from the wildtype protein model and hence were further taken for docking studies. The molecular docking analysis has predicted that these mutations will also be disruptive to the protein-protein interactions between CAMKK2 and PRKAG1 which will create an evident reduction in the kinase activity. The current study has enlightened us that a few of the significant mutations are prime candidates in CAMKK2 which could be the fundamental cause of various bipolar and psychiatric disorders. This is the first detailed study that predicts the deleterious nsSNPs in CAMKK2 and contributes positively in providing a better understanding of disease mechanisms.

Mutational analysis on predicting the impact of high-risk SNPs in human secretary phospholipase A2 receptor (PLA2R1).

PLA2R1 is a transmembrane glycoprotein that acts as an endogenous ligand which stimulates the processes including cell proliferation and cell migration. The SNPs in PLA2R1 is associated with idiopathic membranous nephropathy which is an autoimmune kidney disorder. The present study aimed to explore the structure-function analysis of high risk SNPs in PLA2R1 by using 12 different computational tools. First the functional annotation of SNPs were carried out by sequence based tools which were further subjected to evolutionary conservation analysis. Those SNPs which were predicted as deleterious in both categories were further considered for structure based analysis. The resultant SNPs were C1096S, C545S, C664S, F1257L, F734S, I1174T, I1114T, P177S, P384S, W1198G, W1328G, W692C, W692L, W962R, Y499H. One functional domain of PLA2R1 is already modelled in PDB (6JLI), the full 3D structure of the protein was predicted using I-TASSER homology modelling tool. The stability analysis, structure superimposition, RMSD calculation and docking studies were carried out. The structural analysis predicted four mutations F734S, F1246L, I1174T, W1198G as damaging to the structure of the protein. All these mutations are occurring at the conserved region of CTL domain hence are more likely to abolish the function of the protein. Up to the best of our knowledge, this is the first study that provides in-depth and in-silico analysis of deleterious mutations on structure and function of PLA2R1.

Effects of SLC22A2 (rs201919874) and SLC47A2 (rs138244461) genetic variants on Metformin Pharmacokinetics in Pakistani T2DM patients.

OBJECTIVE: To determine the frequencies of single nucleotide polymorphisms rs201919874 and rs138244461 in genes SLC22A2 and SLC47A2 respectively in Pakistani diabetes patients in order to characterise the genetic variants and determine their association with the pharmacokinetics of metformin. METHODS: The case-control study was conducted at the International Islamic University, Islamabad, Pakistan, from June 2016 to June 2017, and comprised genotypes of diabetic cases and matching controls which were determined following allele-specific polymerase chain reaction. Cases were further divided into Group A and Group B. The former consisted of diabetics who were on monotherapy of metformin, while the latter consisted of diabetics treated with a combination of metformin and sulfonylureas. In-silico analysis was performed to verify the effect of single nucleotide polymorphisms rs201919874 and rs138244461 on the structure of genes. Association was statistically determined using SPSS 18. RESULTS: Of the 1200 subjects, 800(66.6%) were cases and 400(33.3%) were controls. Among the cases, 400(50%) each were in Group A and Group B. Significant difference was observed in the distribution of rs201919874 between Group A and controls (p<0.05) and between Group B and controls (p<0.05) for heterozygous genotypic frequency and for allelic frequency. Conversely, statistically significant difference was observed in rs138244461 (p<0.05) for all genotypic and allelic frequencies. Genotypes were significantly associated with glycated haemoglobin, random and fasting glucose levels in Group A compared to Group B (p<0.05). In-silico analysis showed that both single nucleotide polymorphisms were expected to create significantly damaging structural changes in domains and helix (p<0.05 each). CONCLUSIONS: Both exonic single nucleotide polymorphisms were found to be associated with the pharmacokinetics of metformin.

Computational drug repurposing to predict approved and novel drug-disease associations.

The Drug often binds to more than one targets defined as polypharmacology, one application of which is drug repurposing also referred as drug repositioning or therapeutic switching. The traditional drug discovery and development is a high-priced and tedious process, thus making drug repurposing a popular alternate strategy. We proposed an integrative method based on similarity scheme that predicts approved and novel Drug targets with new disease associations. We combined PPI, biological pathways, binding site structural similarities and disease-disease similarity measures. The results showed 94% Accuracy with 0.93 Recall and 0.94 Precision measure in predicting the approved and novel targets surpassing the existing methods. All these parameters help in elucidating the unknown associations between drug and diseases for finding the new uses for old drugs.

Prediction of HIV Drug Resistance by Combining Sequence and Structural Properties.

Drug resistance is a major obstacle faced by therapist in treating HIV infected patients. The reason behind these phenomena is either protein mutation or the changes in gene expression level that induces resistance to drug treatments. These mutations affect the drug binding activity, hence resulting in failure of treatment. Therefore, it is necessary to conduct resistance testing in order to carry out HIV effective therapy. This study combines both sequence and structural features for predicting HIV resistance by applying SVM and Random Forests classifiers. The model was tested on the mutants of HIV-1 protease and reverse transcriptase. Taken together the features we have used in our method, total contact energies among multiple mutations have a strong impact in predicting resistance as they are crucial in understanding the interactions of HIV mutants. The combination of sequence-structure features offers high accuracy with support vector machines as compared to Random Forests classifier. Both single and acquisition of multiple mutations are important in predicting HIV resistance to certain drug treatments. We have discovered the practicality of these features; hence, these can be used in the future to predict resistance for other complex diseases.

ZK DrugResist 2.0: A TextMiner to extract semantic relations of drug resistance from PubMed.

Extracting useful knowledge from an unstructured textual data is a challenging task for biologists, since biomedical literature is growing exponentially on a daily basis. Building an automated method for such tasks is gaining much attention of researchers. ZK DrugResist is an online tool that automatically extracts mutations and expression changes associated with drug resistance from PubMed. In this study we have extended our tool to include semantic relations extracted from biomedical text covering drug resistance and established a server including both of these features. Our system was tested for three relations, Resistance (R), Intermediate (I) and Susceptible (S) by applying hybrid feature set. From the last few decades the focus has changed to hybrid approaches as it provides better results. In our case this approach combines rule-based methods with machine learning techniques. The results showed 97.67% accuracy with 96% precision, recall and F-measure. The results have outperformed the previously existing relation extraction systems thus can facilitate computational analysis of drug resistance against complex diseases and further can be implemented on other areas of biomedicine.