International Journal of Computational Biology and Drug Design (19 papers in press)
Data Acquisition and Electrical Instrumentation Engineering Modelling for Intelligent Learning and Recognition
by Jun Qin, Yuhao Jiang
Development of interactive computer learning program for genetics and molecular biology applications
by Xiaoli Yang, Bin Chen, Yifan Cai, Charles Tseng
Comparative conformational variation and flexibility analysis of binding site D-loop and its importance in designing of potential tankyrases inhibitors
by Rakesh D. Nimbalkar, Sujit R. Tangadpalliwar, Prabha Garg
Abstract: The principle point of this study is to know the dynamics of D-loop affecting the inhibitory action of tankyrase inhibitors. The knowledge of D-loop conformational variations, flexibility analysis, and its significance in the binding of tankyrase inhibitors (TNKSIs) has not been explored earlier. This study at first focused on observing the conformational changes of D-loop in tankyrases crystal structures. These conformational changes were observed by overlapping of D-loop residues for their open inward and open outward conformation as well as to describe the movement of D-loop towards the neighboring loop (N-loop) or not. Comparative analysis of D-loop of TNKS1 and TNKS2 shows that D-loop of TNKS2 has its conformation more towards the neighboring N-loop. Flexibility analyses of both the isoforms of TNKSs have indicated the most extreme movement of D-loop in the presence of adenosine and nicotinamide site inhibitors. The presence of dual site inhibitor doesn't influence the movement of D-loop significantly. To achieve this, relative residue fluctuation profile examination of undocked and docked complexes were performed. Further, MM/GBSA free energy calculations were used to expose the more prominent binding strength of strong inhibitors and selectivity compared to other respective site inhibitors. Molecular dynamics simulation search revealed that RMSF profile of D-loop of TNKSs affects the binding potential of inhibitors. Thus conformational changes in the D-loop affected by inhibitors can be used to screen the potent tankyrase inhibitors (TNKSIs). The molecular docking, D-loop flexibility analysis, molecular dynamics and MM/GBSA binding energy calculations were carried out.
Keywords: Tankyrase-1; TNKS1; Tankyrase-2; TNKS2; Prime MM/GBSA; molecular docking; Molecular dynamic simulations; D-loop; residue fluctuation.
Interaction of Curcumin with Different Target Proteins of Alzheimers Disease: Docking and MD Simulation Studies
by Shraddha Sriraman, Nanda Gopal Saha, Sujata Roy
Abstract: Curcumin, commonly called turmeric, is a polyphenol derived from the rhizome of the plant, Curcuma longa. Curcumin has been extensively used in the treatment of various medical conditions, including arthritis, cystic fibrosis, and cancer and so on. In addition, curcumin also has a potential role in the prevention and treatment of Alzheimer's disease (AD). Although the effect of curcumin with regards to AD has been studied experimentally, the molecular mechanism is still unknown. Many targets of AD have been identified. In this analysis, the interactions of curcumin with eight different targets of AD have been studied, in order to locate the binding site of curcumin. Based on docking energy, three potential targets such as acetylcholinesterase, cholinesterase and inducible nitric oxide synthase have been selected. Then, MD simulation was performed for those three docked structures. It was found that acetylcholinesterase was the best target of curcumin. Existing experimental results support this finding. The dynamics of interaction at the atomic level was studied to understand the main chemical property of curcumin that can be exploited in treating AD.
Keywords: Alzheimer's disease; Curcumin; Acetylcholinesterase; Molecular Docking and Molecular Dynamics.
An Improved Convex and Concave Index for Revealing the Exposure Degree of Atoms in Protein 3D Structure
by Xiao Wang, Jian Zhao, Yujiao Yan, Jingye Qian, Ping Han, Xiaofeng Song
Abstract: Geometry property of protein surface contributes largely to the protein function in the cells, descriptors of measuring the convexity and concavity of protein surfaces can help to understand the protein function. Motivated by CX algorithm, we developed an improved surface structural parameter named convex-and-concave index (CCI) to describe geometric properties of protein surface. The proposed CCI eliminates the defect of coarse computing for the overlap volume of two adjacent atom spheres in CX algorithm, by dividing the probe sphere into many cubic lattices and labelling the cubic lattices inside the atoms or not respectively. The results indicated that the CCI algorithm improved the accuracy of CX, and not increased the computing complexity. The proposed CCI is a fast and simple method that can accurately describe the exposure degree of atoms in protein and reveal protein functional sites, such as active sites and ubiquitination site.
Keywords: structural parameter; convex-and-concave index; degree of exposure; protein surface; active site; ubiquitination site.
The extravascular penetration of tirapazamine into tumours: a predictive model of the transport and efficacy of hypoxia specific cytotoxic analogues and the potential use of cucurbiturils to facilitate delivery.
by Clifford Fong
Abstract: A multiparameter model of the diffusion, antiproliferative assays IC50 and aerobic and hypoxic clonogenic assays for a wide range of neutral and radical anion forms of tirapazamine (TPZ) analogues has found that: (a) extravascular diffusion is governed by the desolvation, lipophilicity, dipole moment and molecular volume, similar to passive and facilitated permeation through the blood brain barrier and other cellular membranes, (b) hypoxic assay properties of the TPZ analogues show dependencies on the electron affinity, as well as lipophilicity and dipole moment and desolvation, similar to other biological processes involving permeation of cellular membranes, including nuclear membranes, (c) aerobic properties are dependent on the almost exclusively on the electron affinity, consistent with electron transfer involving free radicals being dominant with little or no drug permeation of membranes, and most likely occurring in the extracellular matrix. Application of the model to the DNA binding equilibrium constants of TPZ analogues with acridine or acridine-like moieties show that ligand water desolvation and lipophilicity are the dominant processes governing the DNA intercalation of TPZ analogues. This conclusion is consistent with DFT modelling of the complexes formed by TPZ analogues with neutral and N-protonated acridine moieties which intercalate with the guanine DNA nucleobases. A quantum mechanical study has shown that TPZ can form stable complexes with cucurbituril as a precursor to proof of principle of improved TPZ delivery to tumours.
Keywords: Tirapazamine analogues; intra-tumoural diffusion; anti-cancer; hypoxia; cytotoxicity; DNA binding; cucurbiturils; quantum mechanics.
Structural analysis of protein translocase subunit SecY from Mycobacterium tuberculosis H37Rv: a potential target for anti-tuberculosis drug discovery
by Tilahun Melak Sitote, Sunita Gakkhar
Abstract: Identification of noble drug targets is a very important step in the development of anti-mycobacterial drugs to counter the problem of drug-resistance. The availability of structural information of a specified drug target is one of the druggablity criteria. However, many proteins do not have experimentally solved structure in spite of the efforts of structural genomics projects. In this study, structural analysis on a selected potential drug target of Mycobacterium tuberculosis H37Rv has been carried out. Protein translocase subunit SecY (Rv0732) has been selected since it is a highly ranked potential drug target without solved three-dimensional structure. In silco structural analysis has been carried out to get descriptive three-dimensional structure. The models were generated using Crystal Structure of Secye Translocon from Thermus thermophilus with a Fab Fragment (2ZJS_Y) as a template. The active site has been identified for protein-inhibitor binding.
Keywords: Active site; Drug-resistance tuberculosis; Homology modeling.
Identification of dual target anti-inflammatory inhibitors using merged structure based pharmacophore modelling and docking approach
by Manikandan Selvaraj, Muhd Hanis MD Idris, Siti Norhidayu Mohd Amin, Mohd Zaki Salleh, Teh Lay Kek
Abstract: Merged structure-based pharmacophore modelling followed by 3-D database search and molecular docking were the sequential protocol applied in order to identify selective novel COX-2 and PDE4D as dual target anti-inflammatory inhibitors. Utilization of the key interaction features of crystal structures of COX-2 (pdb: 1CX2) and PDE4D (pdb: N0YN) was exploited as critical component in the selection of dual target inhibitors. Through this approach, nine chalcone and flavones scaffold like compounds were selected as putative dual target anti-inflammatory inhibitors from Asinex database. In general understanding such approach could provide valuable insights into discovery of novel anti-inflammatory inhibitors as therapeutic agents.
Keywords: Cyclooxygenase; phosphodiesterase; dual target; merged pharmacophore; docking.
Adaptive-Fuzzy clustering based texture analysis for classifying liver cancer in abdominal CT images
by Amita Das, Priti Das, S.S. Panda, Sukanta Sabut
Abstract: Segmentation of diseased liver in abdominal CT images is a challenging task due to variations in shapes, tissue similarity between adjoining organs like kidney, heart and also pathologies caused by diseases. The computer aided diagnosis (CAD) systems is very useful in automatic analysis of tumor position and finding a region of interest (ROI) from images. We propose a technique that integrates the fuzzy clustering with adaptive thresholding for segmenting the liver and the finding tumor region in abdominal CT images. Various features like texture features, morphological features and statistical features have been extracted from the output images and used as input to the classifier. The proposed method was evaluated in a series of 45 images taken from MICCAI datasets and open sources. Neural network classifier has been used to classify the malignant and benign tumor of the liver. The efficiency of the proposed algorithm is tested in terms of sensitivity, specificity, and accuracy. The accuracy of 97.82%, 95.74% is achieved in BPN and LVQ and a higher accuracy of 98.82% is achieved with PNN in detecting tumors which are comparable to published results. This method could be an effective solution for identifying the tumor region of liver on CT abdominal images.rnrn
Keywords: CT image; liver; tumor; segmentation; region of interest; neural network classifier.
Comparative analysis of machine learning based QSAR Models and molecular docking studies to screen potential anti-tubercular inhibitors against InhA of Mycobacterium tuberculosis
by Madhulata Kumari
Abstract: Machine learning techniques are advanced computational techniques which can be used to build quantitative structureactivity relationship (QSAR) model of compounds data set to find out important descriptors which are able to predict a specific biological activity from unknown compounds to discover better drugs. In the present study, by optimizing descriptors using correlation-based feature selection, principal component analysis, and genetic programing technique, several machine learning techniques were used to build QSAR models on three different experimental datasets of InhA inhibitors. The best QSAR models were deployed on a data set of 1450 approved drug from drug bank to screen new InhA inhibitors. Amoxicillin was found to show highest predicted activity pIC50=6.54, and Itraconazole was the second compound with highest predicted activity 6.4 (pIC50 ) that was calculated based on the best Random Forest (RF) model using CFS-GS-FW descriptor set in the dataset of ChEMBL997779 of InhA of Mtb. Additionally, screening by molecular docking identified top-ranked ten approved drugs as anti-tubercular hits showing G-scores -8.23 to -6.95 (in kcal/mol) as compared with control compounds(known InhA Mtb inhibitors) G-scores -7.86 to -6.68 (in kcal/mol). Thus results indicate these potent compounds may have the better binding affinity for InhA of Mtb. From ourstudies, we conclude that machine learning based QSAR models can be useful for the development of novel target specific anti-tubercular compounds.
Keywords: Machine learning algorithms; Quantitative structure-activity relationships; Support vector machine; Random forest; Multilayer Perceptron; Genetic Algorithm; Genetic Programming; Regression; Mycobacterium tuberculosis; Gaussian Process; Correlation-based Feature Selection; InhA.
Identifying the dynamic gene regulatory network during latent HIV-1 reactivation using high-dimensional ordinary differential equations
by Jaejoon Song, Michelle Carey, Hongjian Zhu, Hongyu Miao, Juan Ramirez, Hulin Wu
Abstract: Reactivation of latently infected cells have emerged as an important strategy for eradication of HIV. However, genetic mechanisms of regulation after reactivation remains unclear. We describe a five-step pipeline to study the dynamics of the gene regulatory network following a viral reactivation using high-dimensional ordinary differential equations. Our pipeline implements a combination of five different methods, by detecting temporally differentially expressed genes (step 1), clustering genes with similar temporal expression patterns into a small number of gene response modules (step 2), performing a functional enrichment analysis within each gene response module (step 3), identifying a network structure based on the gene response modules using ordinary differential equations (ODE) and a high-dimensional variable selection technique (step 4), and obtaining a gene regulatory model based on refined parameter estimates using nonlinear least squares (step 5). We applied our pipeline to a time course gene expression data of latently infected T-cells following a latency-reversion. We identified 3,926 temporally differentially expressed genes in the latently infected cell relative to the uninfected cell, after viral reactivation. The temporal profiles of these genes were clustered into 95 co-expression patterns. These clusters provide the dynamic gene response modules (set of genes that share similar response patterns over time). A within-module functional annotations were performed using pathway analysis. In addition, we have constructed a regulation network of the dynamic gene response modules using ordinary differential equations and a high-dimensional variable selection technique. Our results indicate that genetic mechanism after viral reactivation of latently infected cells can be described by a regulatory network of gene modules, which consists of genes that share similar temporal expression patterns. Our findings offer new insights for understanding the biological processes underlying the viral cycle after latency.
Keywords: HIV; Gene Regulatory Network; Ordinary Differential Equations.
Special Issue on: ICIBM 2016 Recent Advances in Computational Systems Biology and Bioinformatics
A Flexible Approach to Reconstruct the Genomic Spatial Structure by the Genetic Algorithm
by Yan Zhang, William Hoskins, Ruofan Xia, Xiya Xia, Jim W. Zheng, Jijun Tang
Abstract: The 3D structures of the chromosomes play fundamental roles in essential cellular functions, e.g. gene regulation, gene expression, evolution. HiC technique provides the interaction density between loci on chromosomes.\r\nSeveral approaches have been developed to reconstruct the 3D model of the chromosomes from HiC data. However, all of the approaches are based on a particular mathematical model and lack of flexibility for new development.\r\nWe introduce a novel approach using the genetic algorithm. Our approach is\r\nflexible to accept any mathematical models to build a 3D chromosomal structure.\r\nAlso, our approach outperforms current techniques in accuracy
Keywords: : Genome; Spatial Structure; Genetic Algorithm; HiC.
Signal Translational Efficiency between mRNA Expression and Antibody-based Protein Expression for Breast Cancer and its Subtypes from Cell lines to Tissue
by Aida Yazdanparast, Lang Li, Milan Radovich, Lijun Cheng
Abstract: Background: Although gene transcripts and protein expression have been utilized to classify breast cancer subtypes, it is not clear whether the observed measurement of gene transcript abundance can predict its protein expression. Herein, we attempt to address gene transcript/protein associations using publically-available data on breast cancer tumor tissues and cell lines. Method: Correlation analysis between mRNAs and Reverse-phase protein arrays (RPPA) among 421 primary breast tumors and 33 breast cancer cell lines was conducted. Highly concordant proteins/genes were further analyzed in different breast cancer subtypes. Results: The overall accordance of mRNA/RPPA correlation between cell lines and primary tissue is R2=0.71. Since most of these genes are well known drug targets, highly concordant gene/RPPA associations not only confirm that these gene transcripts can serve as biomarkers for their protein products in drug target selection, but also imply that breast cancer cell lines can serve as good models for primary breast cancer tumors.
Keywords: Breast cancer; Reverse-phase protein array; mRNA; Cell lines; Protein abundance.
Native State of Complement Protein C3d Analysed via Hydrogen Exchange and Conformational Sampling
by Didier Devaurs, Malvina Papanastasiou, Dinler Antunes, Jayvee Abella, Mark Moll, Daniel Ricklin, John Lambris, Lydia Kavraki
Abstract: Hydrogen/deuterium exchange detected by mass spectrometry (HDX-MS) provides valuable information on protein structure and dynamics. Although HDX-MS data is often interpreted using crystal structures, it was suggested that conformational ensembles produced by molecular dynamics simulations yield more accurate interpretations. In this paper, we analyse the complement protein C3d by performing an HDX-MS experiment, and evaluate several interpretation methodologies using an existing prediction model to derive HDX-MS data from protein structure. To interpret and refine C3d's HDX-MS data, we look for a conformation (or conformational ensemble) of C3d that allows computationally replicating this data. We confirm that crystal structures are not a good choice and suggest that conformational ensembles produced by molecular dynamics simulations might not always be satisfactory either. Finally, we show that coarse-grained conformational sampling of C3d produces a conformation from which its HDX-MS data can be replicated and refined.
Keywords: complement protein C3d; hydrogen exchange; mass spectrometry; protein conformational sampling; coarse-grained conformational sampling; native state; X-ray crystallography; molecular dynamics; protein structures; conformational ensembles.
Inhibition of Polyamine biosynthesis for toxicity control in Serratia marcescens strain WW4 by targeting ornithine decarboxylase: A structure-based virtual screening study
by Kalyani Dhusia, Pramod Yadav, Rohit Farmer, Pramod Ramteke
Abstract: Ornithine decarboxylase (ODC) enzyme, catalyzes the decarboxylation of ornithine to form spermidine which is a committed step in the biosynthesis of Polyamines. Polyamines are crucial for growth, cell proliferation and differentiation, but are toxic when produced in excess. Ornithine is the immediate precursor, for the production of polyamines via Polyamine biosynthesis mechanism. In this biosynthesis, ODC plays the central role hence, is considered the key target for inhibitory study. Polyamines being produced by Ornithine, the immediate precursor and ODC) plays the central role in this biosynthesis pathway hence, is considered the key target for inhibitory study. Here, in the present work, structure of ODC was modelled and studied for its active site. 142 Natural products of Indofine Herbal Ingredient from Zinc Database were screened using Autodock Vina for the identification of leading herbal inhibitors. The results obtained from docking showed that Conessine, Sumaresinolic acid, DNC, Exolone, Naringenin, Hesperidin and Baicailin were the top most inhibiting candidates with Docking Affinity -9.7(Kcal/mol), -9.2 (Kcal/mol), -9.0 (Kcal/mol), -8.9 (Kcal/mol), -8.8(Kcal/mol), -8.8(Kcal/mol) and -8.2(Kcal/mol) respectively. According to our findings, Conessine (IUPAC name- N,N-dimethylcon-5-enin-3946;-amine) was found to be the best inhibitor and is an alkaloid which proves its immense importance as metabolites. These herbal inhibitors can turn out to be significantly crucial in controlling the toxicity caused by excess production of polyamines by these PGPBs. Thus, Polyamines being harmful when in excess are necessary to be controlled at their genesis and according to literature, no similar approach has been reported yet in arena of herbal inhibition for polyamine biosynthesis or for toxicity control in PGPBs.
Keywords: Ornithine decarboxylase; Herbal Inhibitor; Molecular dynamics simulation; Docking; Virtual screening.
Risk-associated and pathway-based method to detect association with Alzheimer\'s disease
by Jeffrey Mitchel, Laszlo Prokai, Youping Deng, Fan Zhang, Robert Barber
Abstract: It is becoming increasingly apparent that genes do not function alone but through complex biological pathways in complex diseases such as Alzheimers disease (AD). Unraveling these intricate pathways is essential to understanding biological mechanisms of AD. Pathway-based association analysis allows for the discovery of highly significant pathways from the AD vs normal controls samples. Knowledge of activation of these processes will lead to novel markers identifying their signatures in patients at high risk for AD. Based on the Integrated Pathway Analysis Database (IPAD), we developed pathway-based method to detect association with AD. First, we performed risk associated allele analysis to determine if a major or minor allele is associated with risk. Then we performed pathway-disease association analysis to identify 133 AD-associated pathways. Lastly, we performed pathway-patient association analysis to investigate the patients association and distribution among the 133 pathways. We found 5 AD-associated pathways that have the highest association with patients. We present a pathway-based method to detect AD-associated pathways from GWAS data. Our pathway-based analysis not only provides a technique to identify disease-associated pathways, but also help determine the pathway-patient association. We believe that the method can help us with a comprehensive understanding of the molecular mechanisms underlying complex diseases such as AD.
Keywords: Alzheimer’s disease; pathway analysis; biomarker discovery.
Evaluation of biological and technical variations in low-input RNA-Seq and single-cell RNA-Seq
by Fan Gao, Jae Mun Kim, JiHong Kim, Ming-Yi Lin, Charles Y. Liu, Jonathan J. Russin, Christopher P. Walker, William Mack, Oleg V. Evgrafov, Robert H. Chow, James A. Knowles, Kai Wang
Abstract: Background: Although low-input RNA-Seq and single-cell RNA-Seq have been widely used today, two technical questions remain: (1) in the absence of biological variation, what proportion of technical noise comes from input RNA quantity as compared to bioinformatics tools? (2) in biological samples from single neurons, whether variation in gene expression is attributable to biological heterogeneity or just random noise? To examine the sources of variability, we have generated RNA-Seq data from both low-input RNA (two reference RNA samples at 10pg, 100pg, 1000pg quantity, each with 3-6 replicates) and single neurons (16 and 22 cells from two human brains). Results: We performed comparative analysis of the low-input data using different quantification pipelines and dimensionality reduction algorithms. We also compared functional enrichment of the most variably expressed genes from low-input and single neuron data. In general, the quantity of input RNA is negatively correlated with variation of gene expression from technical replicates. For genes in the medium- and high-expression groups, input RNA amount explains most of the variation, whereas differences in the bioinformatics pipeline explain some variation for the low-expression group. The dimensionality reduction method t-SNE reveals data-inherent aggregation of technical replicates of low-input data, and suggests heterogeneity of single pyramidal neuron transcriptome. Interestingly, the variation in gene expression in single neurons is biologically relevant. Conclusions: We found that variation contributed from bioinformatics pipeline is generally minor compared to the quantity of input RNA. We also demonstrated that t-SNE is more effective than PCA to handle the noises from very low-input RNA-Seq (single-neuron level). All the data sets were made available in public repositories for future benchmarking studies.
Keywords: RNA-Seq; single cell sequencing; bioinformatics; TOPHAT; RSEM; t-SNE; PCA; ANNOVAR; variations.
A GPU-CPU Heterogeneous Algorithm for NGS Read Alignment
by Ahmad Al Kawam, Sunil Khatri, Aniruddha Datta
Abstract: In the Next Generation Sequencing (NGS) read alignment problem, millions of DNA fragments, called reads, are mapped to a reference genome. NGS has unleashed a wealth of genomic information by producing immense amounts of data. It is enabling humanity to learn more about the origins of life and the genetic basis of diseases like cancer. Genomic analysis is typically carried out using traditional computing platforms, which have become a limiting factor in the speed of the process. The massive scale of the problem makes it an attractive target for acceleration. In this paper, we design a read alignment algorithm designed to run on a heterogeneous system composed of a GPU and a multicore CPU. We introduce novel techniques for the alignment process and construct a computational pipeline of overlapped CPU and GPU stages. Our design exploits the GPU's massive parallelism and ability to hide memory access latency to align hundreds of reads concurrently. We use OpenMP to hide I/O latency on a parallel network file system by loading the reads in batches in an overlapped manner (via CPU), processing (mostly via GPU), and writing of separate batches (via CPU) to maximize throughput. We compare our tool with the BWA-mem alignment tool, and the results show substantial speedups.
Keywords: Next Generation Sequencing; Read Alignment; GPU Acceleration.
Neural signature of event-related N200 and P300 modulation in parietal lobe during human response inhibition
by Rupesh Kumar Chikara, Oleksii Komarov, Li-Wei Ko
Abstract: The response inhibition control is important for daily life activities, such as car driving, walking, and playing games. The role of inhibition in many studies remains an issue of debate, most researchers nevertheless agree that some sort of inhibition mechanism is involved in the deliberate cessation of a motor response. Therefore, the stop-signal paradigm has been designed to investigate the response inhibition process. Other aspects also encourage the importance of this study, because stop-signal task performance may contribute to neurological disorders such as schizophrenia disorder and attention deficit hyperactivity disorder (ADHD) or obsessive-compulsive disorder. The aim of this study was to explore EEG modulation during left- and right-hand response inhibitions by using ERP analyses. In this study, we observed inhibition related significant ERP modulation in N200 and P300 waves at frontal, central and parietal regions. These outcomes reveal the response inhibition related neural markers in frontal, central and parietal lobes.From these findings, the statistically independent nature of the inhibition mechanisms of the left-hand and right-hand responses in the frontal, central, and parietal brain areas is clearly marked.
Keywords: Electroencephalography (EEG); Stop-signal task; Response inhibition; Event-related potentials (ERP); N200; P300.