Search

The blood–brain barrier is a major obstacle in treating brain-related disorders, as it does not allow the delivery of drugs into the brain. We developed a method for predicting blood–brain barrier penetrating peptides to facilitate drug delivery into the brain. These blood–brain barrier penetrating peptides (B3PPs) can act as therapeutics, as well as drug delivery agents. We trained, tested, and evaluated our models on blood–brain barrier peptides obtained from the B3Pdb database. First, we computed a wide range of peptide features. Then, we selected relevant peptide features. Finally, we developed numerous machine-learning-based models for predicting blood–brain barrier peptides using the selected features. The random-forest-based model performed the best with respect to the top 80 selected features and achieved a maximal 85.08% accuracy with an AUROC of 0.93. We also developed a webserver, B3pred, that implements our best models. It has three major modules that allow users to predict/design B3PPs and scan B3PPs in a protein sequence.

N‐acetylglucosamine (NAG) belongs to the eight essential saccharides that are required to maintain the optimal health and precise functioning of systems ranging from bacteria to human. In the present study, we have developed a method, NAGbinder, which predicts the NAG‐interacting residues in a protein from its primary sequence information. We extracted 231 NAG‐interacting nonredundant protein chains from Protein Data Bank, where no two sequences share more than 40% sequence identity. All prediction models were trained, validated, and evaluated on these 231 protein chains. At first, prediction models were developed on balanced data consisting of 1,335 NAG‐interacting and noninteracting residues, using various window size. The model developed by implementing Random Forest using binary profiles as the main principle for identifying NAG‐interacting residue with window size 9, performed best among other models. It achieved highest Matthews Correlation Coefficient (MCC) of 0.31 and 0.25, and Area Under Receiver Operating Curve (AUROC) of 0.73 and 0.70 on training and validation data set, respectively. We also developed prediction models on realistic data set (1,335 NAG‐interacting and 47,198 noninteracting residues) using the same principle, where the model achieved MCC of 0.26 and 0.27, and AUROC of 0.70 and 0.71, on training and validation data set, respectively. The success of our method can be appraised by the fact that, if a sequence of 1,000 amino acids is analyzed with our approach, 10 residues will be predicted as NAG‐interacting, out of which five are correct. Best models were incorporated in the standalone version and in the webserver available at https://webs.iiitd.edu.in/raghava/nagbinder/

Fracture studies were carried out on narrow gap welded SA 312 Type 304LN stainless steel straight pipes having circumferential through-wall notch in the weld. Totally six specimens were tested; one specimen was subjected to monotonic loading and the remaining five specimens were subjected to quasi-cyclic loading. Under quasi-cyclic loading, one specimen was tested under incremental displacement-controlled loading and the remaining four specimens were tested constant amplitude load-controlled loading. Based on the load-controlled experimental results, a plot of the cyclic load amplitude versus number of cycles to failure of the specimens was obtained. The experimental results indicate that the piping components subjected to quasi-cyclic loading may fail in very less number of loading cycles even when the load amplitude is sufficiently below the monotonic fracture/collapse load. Crack growth behaviour and cyclic tearing instability in the specimens, which are required for realistic assessment of Leak Before Break applicability, was studied.

InCoB, one of the largest annual bioinformatics conferences in the Asia-Pacific region since its launch in 2002, returned to New Delhi, India after 12 years, with a conference attendance of 314 delegates. The 2018 conference had sessions on Big Data and Algorithms, Next Generation Sequencing and Omics Science, Structure, Function and Interactions, Disease and Drug Discovery and Plant and Agricultural Bioinformatics. The conference also featured an industry track as well as panel discussions on Women in Bioinformatics and Democratization vs. Quality control in academic publishing. Asia Pacific Bioinformatics Interaction & Networking Society (APbians) was launched as an APBionet Special Interest Group. Of the 52 oral presentations made, 22 were accepted in supplemental issues of BMC Bioinformatics, BMC Genomics or BMC Medical Genomics and are briefly reviewed here. Next year’s InCoB will be held in Jakarta, Indonesia from September 10–12, 2019.

Number of beta-lactamase variants have ability to deactivate ceftazidime antibiotic, which is the most commonly used antibiotic for treating infection by Gram-negative bacteria. In this study an attempt has been made to develop a method that can predict ceftazidime resistant strains of bacteria from amino acid sequence of beta-lactamases. We obtained beta-lactamases proteins from the β-lactamase database, corresponding to 87 ceftazidime-sensitive and 112 ceftazidime-resistant bacterial strains. All models developed in this study were trained, tested, and evaluated on a dataset of 199 beta-lactamases proteins. We generate 9149 features for beta-lactamases using Pfeature and select relevant features using different algorithms in scikit-learn package. A wide range of machine learning techniques (like KNN, DT, RF, GNB, LR, SVC, XGB) has been used to develop prediction models. Our random forest-based model achieved maximum performance with AUROC of 0.80 on training dataset and 0.79 on the validation dataset. The study also revealed that ceftazidime-resistant beta-lactamases have amino acids with non-polar side chains in abundance. In contrast, ceftazidime-sensitive beta-lactamases have amino acids with polar side chains and charged entities in abundance. Finally, we developed a webserver “ABCRpred”, for the scientific community working in the era of antibiotic resistance to predict the antibiotic resistance/susceptibility of beta-lactamase protein sequences. The server is freely available at (http://webs.iiitd.edu.in/raghava/abcrpred/).Key PointsCeftazidime is commonly used to treat infection caused by Gram-negative bacteria.Beta-lactamase is responsible for lysing ceftazidime, make it resistant to bacteria.Comparison of resistant and sensitive variants of beta-lactamase.Classification of sensitive and resistant strain of bacteria based on beta-lactamase.Prediction models have been developed using different machine learning techniques.

Background Proinflammatory cytokines are correlated with the severity of disease in patients with COVID-19. IL6-mediated activation of STAT3 proliferates proinflammatory responses that lead to cytokine storm promotion. Thus, STAT3 inhibitors may play a crucial role in managing the COVID-19 pathogenesis. The present study discusses a method for predicting inhibitors against the STAT3 signaling pathway. Method The main dataset comprises 1565 STAT3 inhibitors and 1671 non-inhibitors used for training, testing, and evaluation of models. A number of machine learning classifiers have been implemented to develop the models. Results The outcomes of the data analysis show that rings and aromatic groups are significantly abundant in STAT3 inhibitors compared to non-inhibitors. First, we developed models using 2-D and 3-D chemical descriptors and achieved a maximum AUC of 0.84 and 0.73, respectively. Second, fingerprints are used to build predictive models and achieved 0.86 AUC with an accuracy of 78.70% on the validation dataset. Finally, models were developed using hybrid descriptors, which achieved a maximum of 0.87 AUC with 78.55% accuracy on the validation dataset. Conclusion We used the best model to identify STAT3 inhibitors in FDA-approved drugs and found few drugs (e.g., Tamoxifen and Perindopril) to manage the cytokine storm in COVID-19 patients. A webserver “STAT3In” (https://webs.iiitd.edu.in/raghava/stat3in/) has been developed to predict and design STAT3 inhibitors., Graphical abstract Image 1

In literature, several mRNA, miRNA, lncRNA based biomarkers are identified by genomic analysis to stratify the patients into high and low risk groups of pancreatic adenocarcinoma (PAAD). The identified biomarkers are of limited use in terms of sensitivity and prediction ability. Thus, we aimed to identify the prognostic alternative splicing events and their related mutations in the PAAD. PAAD splicing data of 174 samples (17874 AS events in 6209 genes) and corresponding clinical information was obtained from the SpliceSeq and The Cancer Genome Atlas (TCGA), respectively. Prognostic-index based modeling was used to obtain the best predictive models for the seven AS types. However, model based on multiple spliced events genes (APP; LATS1; MRPL4; LAS1L; STARD10; PHF21A; NMRAL1) outperformed the single event models with a remarkable HR of 9.13 (p-value = 6.42e-10) as well as other existing models. Results from g:Profiler suggest that transcription factors ZF5, ER81, E2F-1/2/3, ER81, Erg, and PEA3 are most related to the prognostic spliced genes. We also identified 565 mutations across 160 spliced genes that have a strong association with the prognostic AS events. The analysis revealed that around 560 of these mutations were not reported before in context to splice event/region. Overall, we conclude that altered AS events may serve as strong indicators for overall survival in pancreatic cancer patients, and novel linkage of the known mutations to the survival-related AS events may provide a new dimension to the advancement of diagnostic and therapeutic interventions in these patients.

Background Allergy is the abrupt reaction of the immune system that may occur after the exposure to allergens such as proteins, peptides, or chemicals. In the past, various methods have been generated for predicting allergenicity of proteins and peptides. In contrast, there is no method that can predict allergenic potential of chemicals. In this paper, we described a method ChAlPred developed for predicting chemical allergens as well as for designing chemical analogs with desired allergenicity. Method In this study, we have used 403 allergenic and 1074 non-allergenic chemical compounds obtained from IEDB database. The PaDEL software was used to compute the molecular descriptors of the chemical compounds to develop different prediction models. All the models were trained and tested on the 80% training data and evaluated on the 20% validation data using the 2D, 3D and FP descriptors. Results In this study, we have developed different prediction models using several machine learning approaches. It was observed that the Random Forest based model developed using hybrid descriptors performed the best, and achieved the maximum accuracy of 83.39% and AUC of 0.93 on validation dataset. The fingerprint analysis of the dataset indicates that certain chemical fingerprints are more abundant in allergens that include PubChemFP129 and GraphFP1014. We have also predicted allergenicity potential of FDA-approved drugs using our best model and identified the drugs causing allergic symptoms (e.g., Cefuroxime, Spironolactone, Tioconazole). Our results agreed with allergenicity of these drugs reported in literature. Conclusions To aid the research community, we developed a smart-device compatible web server ChAlPred ( https://webs.iiitd.edu.in/raghava/chalpred/ ) that allows to predict and design the chemicals with allergenic properties.

Prostate-specific antigen (PSA) is a key biomarker, which is commonly used to screen patients of prostate cancer. There is a significant number of unnecessary biopsies that are performed every year, due to poor accuracy of PSA based biomarker. In this study, we identified alternate biomarkers based on gene expression that can be used to screen prostate cancer with high accuracy. All models were trained and test on gene expression profile of 500 prostate cancer and 51 normal samples. Numerous feature selection techniques have been used to identify potential biomarkers. These biomarkers have been used to develop various models using different machine learning techniques for predicting samples of prostate cancer. Our logistic regression-based model achieved highest AUROC 0.91 with accuracy 82.42% on validation dataset. We introduced a new approach called propensity index, where expression of gene is converted into propensity. Our propensitybased approach improved the performance of classification models significantly and achieved AUROC 0.99 with accuracy 96.36% on validation dataset. We also identified and ranked selected genes which can be used to discriminate prostate cancer patients from health individuals with high accuracy. It was observed that single gene-based biomarkers can only achieve accuracy around 90%. In this study, we got best performance using a panel of 10 genes; random forest model using propensity index.HighlightsApplication of Machine learning techniques to identify Biomarkers for PRAD cancer.Highly accurate models developed for classifying prostate cancer vs. normal sample.Introducing Propensity index concept for enhancing model performance.Top 10 genes identified using feature selection techniques.

Postoperative tumor recurrence is one of the major concerns associated with the poor prognosis of HCC patients. There is yet to elucidate a standard surveillance system for HCC recurrence risk owing to complexity of this malignancy. Generation of multi-omics data from patients facilitate the identification of robust signatures for various diseases. Thus, the current study is an attempt to develop the prognostic models employing multi-omics data to significantly (p-value 12 months). First, we identified key 90RNA, 50miRNA and 50 methylation features and developed prognostic models; attained reasonable performance (C-Index >0.70, HR >2.5), on training and validation datasets. Subsequently, we developed a prognostic (PI) model by integrating the four multi-omics features (SUZ12, hsa-mir-3936, cg18465072, and cg22852503), that are biologically inter-linked with each other. This model achieved reasonable performance on training and validation dataset, i.e. C-Index 0.72, HR of 2.37 (1.61 - 3.50), p-value of 6.72E-06, Brier score 0.19 on training dataset, and C-Index 0.72 (95% CI: 0.63 - 0.80), HR of 2.37 (95% CI: 1.61 - 3.50), p-value of 0.015, Brier score 0.19 on validation dataset. Eventually, Drugbank data was investigated to elucidate therapeutic potential of these signatures. We have identified nine potential drugs against three genes (CA9, IL1A, KCNJ15) that are positively correlated with the tumor recurrence. We anticipate these results from our study will help researchers and clinicians to improve the HCC recurrence surveillance, eventually outcome of patients.

Cancer is one of the most prevailing, deadly and challenging diseases worldwide. The advancement in technology led to the generation of different types of omics data at each genome level that may potentially improve the current status of cancer patients. These data have tremendous applications in managing cancer effectively with improved outcome in patients. This review summarizes the various computational resources and tools housing several types of omics data related to cancer. Major categorization of resources includes—cancer-associated multiomics data repositories, visualization/analysis tools for omics data, machine learning-based diagnostic, prognostic, and predictive biomarker tools, and data analysis algorithms employing the multiomics data. The review primarily focuses on providing comprehensive information on the open-source multiomics tools and data repositories, owing to their broader applicability, economic-benefit and usability. Sections including the comparative analysis, tools applicability and possible future directions have also been discussed in detail. We hope that this information will significantly benefit the researchers and clinicians, especially those with no sound background in bioinformatics and who lack sufficient data analysis skills to interpret something from the plethora of cancer-specific data generated nowadays.

In this study, we attempted to identify prognostic biomarkers for predicting survival risk of uterine corpus endometrial cancer (UCEC) patients from the gene expression profile of pattern recognition receptors (PRRs). A wide range of feature selection techniques have been tried, including network-based methods to identify a small number of genes from 331 PRR genes. Firstly, a risk stratification model has been developed using biomarker genes selected using a network-based approach and achieved HR=1.37 with p=0.294. Secondly, we developed a risk stratification model using biomarker of seven genes obtained from clustering and achieved HR=9.14 and p= 1.49×10-12. Finally, we developed various combinatorial models using biomarker of 15 PRR genes that were significantly associated with UCEC survival. We found that a multiple genes-based risk stratification model using nine genes (CLEC1B, CLEC3A, IRF7, CTSB, FCN1, RIPK2, NLRP10, NLRP9 and SARM1) gave the best result (HR=10.70, p=1.1×10-12, C=0.76, log-rank-p=8.15×10-14). The performance of this model improved significantly when we used the clinical stage of patients in combination with the expression of nine genes and achieved HR=15.23 (p=2.21×10-7, C=0.78, log-rank-p=2.76×10-17). We also developed classification models that can classify high-risk patients (survive ≤ 4.3 years) and low-risk patients (survive > 4.3 years) and achieved AUROC of 0.86. It was observed that specific genes are positively correlated with overall survival of UCEC patients. Based on these observations, we identified potential immunotherapeutic agents for treating UCEC patients.

ObjectivesAberrant expression of apoptotic genes has been associated with papillary thyroid carcinoma (PTC) in the past, however, their prognostic role and utility as biomarkers remains poorly understood.Materials and methodsIn this study, we analysed 505 PTC patients by employing Cox-PH regression techniques, prognostic index models and machine learning methods to elucidate the relationship between overall survival (OS) of PTC patients and 165 apoptosis related genes.ResultsIt was observed that nine genes (ANXA1, TGFBR3, CLU, PSEN1, TNFRSF12A, GPX4, TIMP3, LEF1, BNIP3L) showed significant association with OS of PTC patients. Five out of nine genes were found to be positively correlated with OS of the patients, while the remaining four genes were negatively correlated. These genes were used for developing risk prediction models. Our voting-based model achieved highest performance (HR=41.59, p=3.36×10−4, C=0.84, logrank-p=3.8×10−8). The performance of voting-based model improved significantly when we used the age of patients with prognostic biomarker genes and achieved HR=57.04 with p=10−4 (C=0.88, logrank-p=1.44×10−9). We also developed classification models that can classify high risk patients (survival ≤ 6 years) and low risk patients (survival > 6 years). Our best model achieved AUROC of 0.92. Since these genes can also be used as potential therapeutic targets in PTC, we identified potential drug molecules which could modulate their expression profile.ConclusionThis study briefly revealed the key prognostic biomarker genes in the apoptotic pathway whose altered expression is associated with PTC progression and aggressiveness. In addition to this, risk assessment models proposed here can help in efficient management of PTC patients.

A web-based resource CoronaVIR (https://webs.iiitd.edu.in/raghava/coronavir/) has been developed to maintain the predicted and existing information on coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). We have integrated multiple modules, including "Genomics," "Diagnosis," "Immunotherapy," and "Drug Designing" to understand the holistic view of this pandemic medical disaster. The genomics module provides genomic information of different strains of this virus to understand genomic level alterations. The diagnosis module includes detailed information on currently-in-use diagnostics tests as well as five novel universal primer sets predicted using in silico tools. The Immunotherapy module provides information on epitope-based potential vaccine candidates (e.g., LQLPQGTTLPKGFYA, VILLNKHIDAYKTFPPTEPKKDKKKK, EITVATSRTLS, GKGQQQQGQTV, SELVIGAVILR) predicted using state-of-the-art software and resources in the field of immune informatics. These epitopes have the potential to activate both adaptive (e.g., B cell and T cell) and innate (e.g., vaccine adjuvants) immune systems as well as suitable for all strains of SARS-CoV-2. Besides, we have also predicted potential candidates for siRNA-based therapy and RNA-based vaccine adjuvants. The drug designing module maintains information about potential drug targets, tertiary structures, and potential drug molecules. These potential drug molecules were identified from FDA-approved drugs using the docking-based approach. We also compiled information from the literature and Internet on potential drugs, repurposing drugs, and monoclonal antibodies. To understand host-virus interaction, we identified cell-penetrating peptides in the virus. In this study, state-of-the-art techniques have been used for predicting the potential candidates for diagnostics and therapeutics.

The blood-brain barrier poses major hurdles in the treatment of brain-related ailments. Over the past decade, interest in peptides-based therapeutics has thrived a lot because of their higher benefit to risk ratio. However, a complete knowledgebase providing a well-annotated picture of the peptide as a therapeutic molecule to cure brain-related ailments is lacking. We have built up a knowledgebase B3Pdb on blood-brain barrier (BBB)-penetrating peptides in the present study. The B3Pdb holds clinically relevant experimental information on 1225 BBB-penetrating peptides, including mode of delivery, animal model, in vitro/in vivo experiments, chemical modifications, length. Hoping that drug delivery systems can improve central nervous system disorder-related therapeutics. In this regard, B3Pdb is an important resource to support the rational design of therapeutics peptides for CNS-related disorders. The complete ready-to-use and updated database with a user-friendly web interface is available to the scientific community at https://webs.iiitd.edu.in/raghava/b3pdb/ .

Interleukin 6 (IL-6) is a pro-inflammatory cytokine that stimulates acute phase responses, hematopoiesis and specific immune reactions. Recently, it was found that the IL-6 plays a vital role in the progression of COVID-19, which is responsible for the high mortality rate. In order to facilitate the scientific community to fight against COVID-19, we have developed a method for predicting IL-6 inducing peptides/epitopes. The models were trained and tested on experimentally validated 365 IL-6 inducing and 2991 non-inducing peptides extracted from the immune epitope database. Initially, 9149 features of each peptide were computed using Pfeature, which were reduced to 186 features using the SVC-L1 technique. These features were ranked based on their classification ability, and the top 10 features were used for developing prediction models. A wide range of machine learning techniques has been deployed to develop models. Random Forest-based model achieves a maximum AUROC of 0.84 and 0.83 on training and independent validation dataset, respectively. We have also identified IL-6 inducing peptides in different proteins of SARS-CoV-2, using our best models to design vaccine against COVID-19. A web server named as IL-6Pred and a standalone package has been developed for predicting, designing and screening of IL-6 inducing peptides (https://webs.iiitd.edu.in/raghava/il6pred/).

Globally fermented foods are in demands due to their functional and nutritional benefits. These foods are sources of probiotic organisms and bioactive peptides, various amino acids, enzymes etc. that provides numerous health benefits. FermFooDb (https://webs.iiitd.edu.in/raghava/fermfoodb/) is a manually curated database of bioactive peptides derived from wide range of foods that maintain comprehensive information about peptides and process of fermentation. This database comprises of 2205 entries with following major fields, peptide sequence, Mass and IC50, food source, functional activity, fermentation conditions, starter culture, testing conditions of sequences in vitro or in vivo, type of model and method of analysis. The bioactive peptides in our database have wide range of therapeutic potentials that includes antihypertensive, ACE-inhibitory, antioxidant, antimicrobial, immunomodulatory and cholesterol lowering peptides. These bioactive peptides were derived from different types of fermented foods that include milk, cheese, yogurt, wheat and rice. Numerous, web-based tools have been integrated to retrieve data, peptide mapping of proteins, similarity search and multiple-sequence alignment. This database will be useful for the food industry and researchers to explore full therapeutic potential of fermented foods from specific cultures., Bioactive peptide, Database, Webserver, Microbial fermentation, Functional food.

Risk assessment in cutaneous melanoma (CM) patients is one of the major challenges in the effective treatment of CM patients. Traditionally, clinico-pathological features such as Breslow thickness, American Joint Committee on Cancer (AJCC) tumor staging, etc. are utilized for this purpose. However, due to advancements in technology, most of the upcoming risk prediction methods are gene-expression profile (GEP) based. In this study, we have tried to develop new GEP and clinico-pathological features-based biomarkers and assessed their prognostic strength in contrast to existing prognostic methods. We developed risk prediction models using the expression of the genes associated with different cancer-related pathways and got a maximum hazard ratio (HR) of 2.52 with p-value ~10−8 for the apoptotic pathway. Another model, based on combination of apoptotic and notch pathway genes boosted the HR to 2.57. Next, we developed models based on individual clinical features and got a maximum HR of 2.45 with p-value ~10−6 for Breslow thickness. We also developed models using the best features of clinical as well as gene-expression data and obtained a maximum HR of 3.19 with p-value ~10−9. Finally, we developed a new ensemble method using clinical variables only and got a maximum HR of 6.40 with p-value ~10−15. Based on this method, a web-based service and an android application named ‘CMcrpred’ is available at (https://webs.iiitd.edu.in/raghava/cmcrpred/) and Google Play Store respectively to facilitate scientific community. This study reveals that our new ensemble method based on only clinico-pathological features overperforms methods based on GEP based profiles as well as currently used AJCC staging. It also highlights the need to explore the full potential of clinical variables for prognostication of cancer patients., Bioinformatics; Cancer research; Genetics; Oncology; Cancer; Prognosis; Melanoma; Survival analysis; Skin; Risk prediction.

Globally, ~20% of cancer malignancies are associated with virus infections. Lung cancer is the most prevalent cancer and has a 10% 5-year survival rate when diagnosed at stage IV. Cancer vaccines and oncolytic immunotherapy are promising treatment strategies for better clinical outcomes in advanced-stage cancer patients. Here, we used a reverse vaccinology approach to devise subunit vaccine candidates against lung cancer-causing oncogenic viruses. Protein components (945) from nine oncogenic virus species were systematically analyzed to identify epitope-based subunit vaccine candidates. Best vaccine candidates were identified based on their predicted ability to stimulate humoral and cell-mediated immunity and avoid self-tolerance. Using a rigorous integrative approach, we identified 125 best antigenic epitopes with predicted B-cell, T-cell, and/or MHC-binding capability and vaccine adjuvant potential. Thirty-two of these antigenic epitopes were predicted to have IL-4/IFN-gamma inducing potential and IL-10 non-inducing potential and were predicted to bind 15 MHC-type I and 49 MHC-type II alleles. All 32 epitopes were non-allergenic and 31 were non-toxic. The identified epitopes showed good conservancy and likely bind a broad class of human HLA alleles, indicating promiscuous potential. The majority of best antigenic epitopes were derived from Human papillomavirus and Epstein-Barr virus proteins. Of the 32 epitopes, 25 promiscuous epitopes were related to E1 and E6 envelope genes and were present in multiple viral strains/species, potentially providing heterologous immunity. Further validating our results, 38 antigenic epitopes were also present in the largest experimentally-validated epitope resource, Immune Epitope Database and Analysis Resource. We further narrowed the selection to 29 antigenic epitopes with the highest immunogenic/immune-boosting potential. These epitopes possess tremendous therapeutic potential as vaccines against lung cancer-causing viruses and should be validated in future experiments. All findings are available at https://webs.iiitd.edu.in/raghava/vlcvirus/.

AlgPred 2.0 is a web server developed for predicting allergenic proteins and allergenic regions in a protein. It is an updated version of AlgPred developed in 2006. The dataset used for training, testing and validation consists of 10 075 allergens and 10 075 non-allergens. In addition, 10 451 experimentally validated immunoglobulin E (IgE) epitopes were used to identify antigenic regions in a protein. All models were trained on 80% of data called training dataset, and the performance of models was evaluated using 5-fold cross-validation technique. The performance of the final model trained on the training dataset was evaluated on 20% of data called validation dataset; no two proteins in any two sets have more than 40% similarity. First, a Basic Local Alignment Search Tool (BLAST) search has been performed against the dataset, and allergens were predicted based on the level of similarity with known allergens. Second, IgE epitopes obtained from the IEDB database were searched in the dataset to predict allergens based on their presence in a protein. Third, motif-based approaches like multiple EM for motif elicitation/motif alignment and search tool have been used to predict allergens. Fourth, allergen prediction models have been developed using a wide range of machine learning techniques. Finally, the ensemble approach has been used for predicting allergenic protein by combining prediction scores of different approaches. Our best model achieved maximum performance in terms of area under receiver operating characteristic curve 0.98 with Matthew’s correlation coefficient 0.85 on the validation dataset. A web server AlgPred 2.0 has been developed that allows the prediction of allergens, mapping of IgE epitope, motif search and BLAST search (https://webs.iiitd.edu.in/raghava/algpred2/).

This article reviews more than 50 computational resources developed in past two decades for forecasting of antibiotic resistance (AR)-associated mutations, genes and genomes. More than 30 databases have been developed for AR-associated information, but only a fraction of them are updated regularly. A large number of methods have been developed to find AR genes, mutations and genomes, with most of them based on similarity-search tools such as BLAST and HMMER. In addition, methods have been developed to predict the inhibition potential of antibiotics against a bacterial strain from the whole-genome data of bacteria. This review also discuss computational resources that can be used to manage the treatment of AR-associated diseases.

Increasing use of therapeutic peptides for treating cancer has received considerable attention of the scientific community in the recent years. The present study describes thein silicomodel developed for predicting and designing anticancer peptides (ACPs). ACPs residue composition analysis revealed the preference of A, F, K, L and W. Positional preference analysis revealed that residue A, F and K are preferred at N-terminus and residue L and K are preferred at C-terminus. Motif analysis revealed the presence of motifs like LAKLA, AKLAK, FAKL, LAKL in ACPs. Prediction models were developed using various input features and implementing different machine learning classifiers on two datasets main and alternate dataset. In the case of main dataset, ETree Classifier based model developed using dipeptide composition achieved maximum MCC of 0.51 and 0.83 AUROC on the training dataset. In the case of alternate dataset, ETree Classifier based model developed using amino acid composition performed best and achieved the highest MCC of 0.80 and AUROC of 0.97 on the training dataset. Models were trained and tested using five-fold cross validation technique and their performance was also evaluated on the validation dataset. Best models were implemented in the webserver AntiCP 2.0, freely available athttps://webs.iiitd.edu.in/raghava/anticp2. The webserver is compatible with multiple screens such as iPhone, iPad, laptop, and android phones. The standalone version of the software is provided in the form of GitHub package as well as in docker technology.

Dengue virus (DENV) researchers often face challenges with the highly time-consuming process of collecting and curating information on known inhibitors during the standard drug discovery process. To this end, however, required collective information is not yet available on a single platform. Hence, we have developed the DenvInD database for experimentally validated DENV inhibitors against its known targets presently hosted at https://webs.iiitd.edu.in/raghava/denvind/. This database provides comprehensive information, i.e. PubChem IDs, SMILES, IC50, EC50, CC50, and wherever available Ki values of the 484 compounds in vitro validated as inhibitors against respective drug targets of DENV. Also, the DenvInD database has been linked to the user-friendly web-based interface and accessibility features, such as simple search, advanced search and data browsing. All the required data curation was conducted manually from the reported scientific literature and PubChem. The collected information was then organized into the DenvInD database using sequence query language under user interface by hypertext markup language. DenvInD is the first useful repository of its kind which would augment the DENV drug discovery research by providing essential information on known DENV inhibitors for molecular docking, computational screening, pharmacophore modeling and quantitative structure-activity relationship modeling.

CancerEnD is an integrated resource developed for annotating 8524 unique expressed enhancers, associated genes, somatic mutations and copy number variations of 8063 cancer samples from 18 cancer types of TCGA. Somatic mutation data was taken from the COSMIC repository. To delineate the relationship of change in copy number of enhancer elements with the prognosis of cancer patients, survival analysis was done using the survival package in R. We identified 1762 overall survival associated enhancers, which can be used for prognostic purposes of cancer patients in a tissue-specific manner. CancerEnD (https://webs.iiitd.edu.in/raghava/cancerend/) is developed on a user-friendly responsive template, that enables searching, browsing and downloading of the annotated enhancer elements in terms of gene expression, copy number variation and survival association. We hope it provides a promising avenue for researchers to facilitate the understanding of enhancer deregulation in tumorigenesis, and to identify new biomarkers for therapy and disease-diagnosis.

Liver cancer is the fourth major lethal malignancy worldwide. To understand the development and progression of liver cancer, biomedical research generated a tremendous amount of transcriptomics and disease-specific biomarker data. However, dispersed information poses pragmatic hurdles to delineate the significant markers for the disease. Hence, a dedicated resource for liver cancer is required that integrates scattered multiple formatted datasets and information regarding disease-specific biomarkers. Liver Cancer Expression Resource (CancerLivER) is a database that maintains gene expression datasets of liver cancer along with the putative biomarkers defined for the same in the literature. It manages 115 datasets that include gene-expression profiles of 9611 samples. Each of incorporated datasets was manually curated to remove any artefact; subsequently, a standard and uniform pipeline according to the specific technique is employed for their processing. Additionally, it contains comprehensive information on 594 liver cancer biomarkers which include mainly 315 gene biomarkers or signatures and 178 protein- and 46 miRNA-based biomarkers. To explore the full potential of data on liver cancer, a web-based interactive platform was developed to perform search, browsing and analyses. Analysis tools were also integrated to explore and visualize the expression patterns of desired genes among different types of samples based on individual gene, GO ontology and pathways. Furthermore, a dataset matrix download facility was provided to facilitate the users for their extensive analysis to elucidate more robust disease-specific signatures. Eventually, CancerLivER is a comprehensive resource which is highly useful for the scientific community working in the field of liver cancer.Availability: CancerLivER can be accessed on the web at https://webs.iiitd.edu.in/raghava/cancerliver.

In the present study, a systematic effort has been made to predict the hemolytic potency of chemically modified peptides. All models have been trained, tested, and evaluated on a dataset that contains 583 modified hemolytic peptides and a balanced number of non-hemolytic peptides. Machine learning techniques have been used to build the classification models using an immense range of peptide features that include 2D, 3D descriptors, fingerprints, atom, and diatom compositions. Random Forest based model developed using fingerprints as an input feature achieved maximum accuracy of 78.33% with AUC of 0.86 on the main dataset and accuracy of 78.29% with AUC of 0.85 on the validation dataset. Models developed in this study have been incorporated in a web server “HemoPImod” to facilitate the scientific community (http://webs.iiitd.edu.in/raghava/hemopimod/).

One of the emerging technologies to fight against cancer is oncolytic virus-based immunotherapy. Recently, the FDA approved an oncolytic virus T-vec for the treatment of melanoma. To facilitate the scientific community, we build a manually-curated repository of oncolytic viruses called OvirusTdb (https://webs.iiitd.edu.in/raghava/ovirustdb/). The repository maintains comprehensive information on therapeutically important oncolytic viruses with 5927 records where each record has 25 fields such as the virus species, cancer cell line, synergism with anti-cancer drugs, and many more. It stores information on 09 types of DNA, 15 types of RNA; 300 recombinant and 09 wild-type viral strains; tested against 124 cancer types and 427 cancer cell lines. Approximately, 1047 records suggest improved anti-cancer response using the combinatorial approach with chemotherapeutic agents. Nearly, 3243 and 1506 records indicate cancer cell death via apoptosis induction and immune activation, respectively. OvirusTdb may facilitate researchers in designing and discovering new oncolytic viruses for effective cancer treatment.

IndiaBioDb (https://webs.iiitd.edu.in/raghava/indiabiodb/) is a manually curated comprehensive repository of bioinformatics resources developed and maintained by Indian researchers. This repository maintains information about 543 freely accessible functional resources that include around 258 biological databases. Each entry provides a complete detail about a resource that includes the name of resources, web link, detail of publication, information about the corresponding author, name of institute, type of resource. A user-friendly searching module has been integrated, which allows users to search our repository on any field. In order to retrieve categorized information, we integrate the browsing facility in this repository. This database can be utilized for extracting the useful information regarding the present scenario of bioinformatics inclusive of all research labs funded by government and private bodies of India. In addition to web interface, we also developed mobile to facilitate the scientific community.

IntroductionRecently, the rise in the incidences of thyroid cancer worldwide renders it to be the sixth most common cancer among women. Commonly, Fine Needle Aspiration biopsy predominantly facilitates the diagnosis of the nature of thyroid nodules. However, it is inconsiderable in determining the tumor's state, i.e., benign or malignant. This study aims to identify the key RNA transcripts that can segregate the early and late-stage samples of Thyroid Carcinoma (THCA) using RNA expression profiles.Materials and methodsIn this study, we used the THCA RNA-Seq dataset of The Cancer Genome Atlas, consisting of 500 cancer and 58 normal (adjacent non-tumorous) samples obtained from the Genomics Data Commons (GDC) data portal. This dataset was dissected to identify key RNA expression features using various feature selection techniques. Subsequently, samples were classified based on selected features employing different machine learning algorithms.ResultsSingle gene ranking based on the Area Under the Receiver Operating Characteristics (AUROC) curve identified the DCN transcript that can classify the early-stage samples from late-stage samples with 0.66 AUROC. To further improve the performance, we identified a panel of 36 RNA transcripts that achieved F1 score of 0.75 with 0.73 AUROC (95% CI: 0.62-0.84) on the validation dataset. Moreover, prediction models based on 18-features from this panel correctly predicted 75% of the samples of the external validation dataset. In addition, the multiclass model classified normal, early, and late-stage samples with AUROC of 0.95 (95% CI: 0.84-1), 0.76 (95% CI: 0.66-0.85) and 0.72 (95% CI: 0.61-0.83) on the validation dataset. Besides, a five protein-coding transcripts panel was also recognized, which segregated cancer and normal samples in the validation dataset with F1 score of 0.97 and 0.99 AUROC (95% CI: 0.91-1).ConclusionWe identified 36 important RNA transcripts whose expression segregated early and late-stage samples with reasonable accuracy. The models and dataset used in this study are available from the webserver CancerTSP (http://webs.iiitd.edu.in/raghava/cancertsp/).

One of the fundamental challenges in designing drug molecule against a disease target or protein is to predict binding affinity between target and drug or small molecule. In this review, our focus will be on advancement in the field of protein-small molecule interaction. This review has been divided into four major sections. In the first section, we will cover software developed for protein structure prediction. This will include prediction of binding pockets and post-translation modifications in proteins. In the second section, we will discuss software packages developed for predicting small-molecule interacting residues in a protein. Advances in the field of docking particularly advancement in the knowledgebased force fields will be discussed in the third part of the review. This section will also cover the method developed for predicting affinity between protein and drug molecules. The fourth section of the review will describe miscellaneous techniques used for designing drug molecules, like pharmacophore modelling. Our major emphasis in this review will be on computational tools that are available free for academic use.

CancerPDF (Cancer Peptidome Database of bioFluids) is a comprehensive database of endogenous peptides detected in the human biofluids. The peptidome patterns reflect the synthesis, processing and degradation of proteins in the tissue environment and therefore can act as a gold mine to probe the peptide-based cancer biomarkers. Although an extensive data on cancer peptidome has been generated in the recent years, lack of a comprehensive resource restrains the facility to query the growing community knowledge. We have developed the cancer peptidome resource named CancerPDF, to collect and compile all the endogenous peptides isolated from human biofluids in various cancer profiling studies. CancerPDF has 14,367 entries with 9,692 unique peptide sequences corresponding to 2,230 unique precursor proteins from 56 high-throughput studies for ~27 cancer conditions. We have provided an interactive interface to query the endogenous peptides along with the primary information such as m/z, precursor protein, the type of cancer and its regulation status in cancer. To add-on, many web-based tools have been incorporated, which comprise of search, browse and similarity identification modules. We consider that the CancerPDF will be an invaluable resource to unwind the potential of peptidome-based cancer biomarkers. The CancerPDF is available at the web address http://crdd.osdd.net/raghava/cancerpdf/.

Human Leukocyte Antigen (HLA) is an essential component of the immune system which stimulates immune cells to provide protection and defense against cancer. More than thousands of HLA alleles have been reported in the literature; but, only a specific set of HLA alleles expressed in an individual. Recognition of cancer-associated mutations by the immune system depends on the presence of a particular set of alleles, that elicit an immune response to fight against cancer. It indicates that the occurrence of specific HLA alleles also affects the outcome of the cancer patients. In the current study, prediction models have been developed using 415 skin cutaneous melanoma (SKCM) patients for predicting the overall survival of patients from their HLA-alleles. It has been observed that, the presence of certain superalleles in the patients, is responsible for improved overall survival which were referred as favourable superalleles like HLA-B*55 (HR=0.15, 95% CI 0.034 to 0.67), HLA-A*01 (HR=0.5, 95% CI 0.3 to 0.8). In contrast, presence of certain superalleles in the patients is responsible for their poor survival, those superalleles were referred as unfavourable superalleles such as HLA-B*50 (HR=2.76, 95% CI 1.284 to 5.941), HLA-DRB1*12 (HR=3.44, 95% CI 1.64 to 7.2). We developed prediction models using 14 HLA-superalleles and five clinical characteristics for predicting high-risk SKCM patients and achieve HR=4.52 (95% CI 3.088-6.609) with p-value = 8.01E-15. Lastly, we provide a web-based service to community for predicting the risk in SKCM patients (https://webs.iiitd.edu.in/raghava/skcmhrp/)

1.AbstractThis study describes a method developed for predicting pattern recognition receptors (PRRs), which are an integral part of the immune system. The models developed here were trained and evaluated on the largest possible non-redundant PRRs, and non-pattern recognition receptors (Non-PRRs) obtained from PRRDB 2.0. Firstly, a similarity-based approach using BLAST was used to predict PRRs and got limited success due to a large number of no-hits. Secondly, machine learning-based models were developed using sequence composition and achieved a maximum MCC of 0.63. In addition to this, models were developed using evolutionary information in the form of PSSM composition and achieved maximum MCC value of 0.66. Finally, we developed hybrid models that combined a similarity-based approach using BLAST and machine learning-based models. Our best model, which combined BLAST and PSSM based model, achieved a maximum MCC value of 0.82 with an AUROC value of 0.95, utilizing the potential of both similarity-based search and machine learning techniques. In order to facilitate the scientific community, we also developed a web server “PRRpred” based on the best model developed in this study (http://webs.iiitd.edu.in/raghava/prrpred/).

BackgroundIn past number of web-based resources has been developed in the field of Bioinformatics. These resources are heavily used by scientific community to provide solution for challenges faced by experimental researchers particularly in the field of biomedical sciences. There are number of challenges in utilizing full potential of these services that includes internet speed, limits on computing power, and security of data. In order to enhance utilities of these web-based assets, we developed a docker-based container that integrates large number resources available in literature.ResultsThis paper describes GPSRdocker a docker-based container developed for providing wide-range of computational tools in the field of bioinformatics particularly in genomics, proteomics and system biology. Majority of tools integrated in GPSRdocker are based on web services developed at Raghava’s group in last two decades. Broadly, these tools can be categorized in three categories; i) general scripts, ii) supporting software and iii) major standalone software. In order to facilitate students or developers working in the field of bioinformatics, we developed general scripts in Perl and Python. These general-purpose scripts serve as building block for any bioinformatics tools like computing features/descriptors of a protein. Supporting software packages includes SCIKIT, WEKA, SVMlight, and PSI-BLAST; these software packages allow one to develop/implement bioinformatics software. Major Standalone software is core of this container which allows predicting function/class of biomolecules. These tools can be classified broadly in following categories; protein annotation, epitope-based vaccines, prediction of interaction and drug discovery.ConclusionA docker-based container has been developed which can be easily run on any operating system as well as it can be directly ported on cloud. Scripts can be run to build pipelines for addressing problems at system level like prediction of vaccine candidate for a pathogen. GPSRdocker including manual is available free for academic use from https://webs.iiitd.edu.in/gpsrdocker.

Metastatic state of the Skin Cutaneous Melanoma (SKCM) has led to high mortality rate worldwide. Previously, various studies have revealed the association of the metastatic melanoma with the diminished survival rate in comparison to primary tumors. Thus, prediction of melanoma at primary tumor state is crucial to employ optimal therapeutic strategy for prolonged survival of patients. The RNA, miRNA and methylation data of The Cancer Genome Atlas (TCGA) cohort of SKCM is comprehensively analysed to recognize key genomic features that can categorize various states of metastatic tumors from primary tumors with high precision. Subsequently, various prediction models were developed using filtered genomic features implementing various machine learning techniques to classify these primary tumors from metastatic tumors. The SVC model (with class weight and RBF kernel) developed using 17 mRNA features achieved maximum MCC 0.73 with sensitivity, specificity and accuracy 89.19%, 90.48% and 89.47% respectively on independent validation dataset. Our study reveals that gene expression based features performs better than features obtained from miRNA profiling and epigenomic profiling. Our analysis shows that the expression of genesC7, MMP3, KRT14, KRT17, MASP1, and miRNA hsa-mir-205 and hsa-mir-203a are among the key genomic features that may substantially contribute to the oncogenesis of melanoma even on the basis of simple expression threshold. The major prediction models and analysis modules to predict metastatic and primary tumor samples of SKCM are available from a webserver, CancerSPP (http://webs.iiitd.edu.in/raghava/cancerspp/).

The high mortality rate of hepatocellular carcinoma (HCC) is primarily due to its late diagnosis. In the past numerous attempts have been made to design genetic biomarkers for the identification of HCC; unfortunately, most of the studies are based on a small dataset obtained from a specific platform or lack of their reasonable validation performance on the external datasets. In order to identify a universal expression-based diagnostic biomarker panel for HCC that can be applicable across multiple platforms; we have employed large scale transcriptomic profiling datasets containing a total of 2,306 HCC and 1,655 non-tumorous tissue samples. These samples were obtained from 29 studies generated by mainly four types of profiling techniques include Affymetrix, Illumina, Agilent and High-throughput-seq, which implemented a wide range of platforms. Firstly, we scrutinized 26 genes that are differentially expressed or regulated in uniform pattern among numerous datasets. Subsequently, we identified three genes (FCN3, CLEC1B, & PRC1) panel-based HCC biomarker using different machine learning techniques include Simple-threshold based approach, Extra Trees, Support Vector Machine, Random Forest, K Neighbors Classifier, Logistic Regression etc. Three-genes panel-based HCC biomarker classified HCC samples and non-tumorous samples of training and three external validation datasets with an accuracy between 93 to 98% and AUROC (Area Under Receiver Operating Characteristic curve) in a range of 0.97 to 1.0. Furthermore, the prognostic potential of these genes was evaluated on TCGA cohort and GSE14520 cohort using univariate survival analysis revealed that these genes are independent prognostic indicators for various types of the survivals, i.e. OS (Overall Survival), PFS (Progression-Free Survival), DFS/RFS (Disease-Free Survival/Recurrence-Free Survival) and DSS (Disease-Specific Survival) of HCC patients and significantly stratify high-risk and low-risk HCC patients (p-value http://webs.iiitd.edu.in/raghava/hccpred/).

Understanding the gene regulatory network governing cancer initiation and progression is necessary, although it remains largely unexplored. Enhancer elements represent the center of this regulatory circuit. The study aims to identify the gene expression change driven by copy number variation in enhancer elements of pancreatic adenocarcinoma (PAAD). The pancreatic tissue specific enhancer and target gene data were taken from EnhancerAtlas. The gene expression and copy number data were taken from The Cancer Genome Atlas (TCGA). Differentially expressed genes (DEGs) and copy number variations (CNVs) were identified between matched tumor-normal samples of PAAD. Significant CNVs were matched onto enhancer coordinates by using genomic intersection functionality from BEDTools. By combining the gene expression and CNV data, we identified 169 genes whose expression shows a positive correlation with the CNV of enhancers. We further identified 16 genes which are regulated by a super enhancer and 15 genes which have high prognostic potential (Z-score &gt, 1.96). Cox proportional hazard analysis of these genes indicates that these are better predictors of survival. Taken together, our integrative analytical approach identifies enhancer CNV-driven gene expression change in PAAD, which could lead to better understanding of PAAD pathogenesis and to the design of enhancer-based cancer treatment strategies.

BackgroundNanostructures generated by self-assembly of peptides yield nanomaterial that has many therapeutic applications, including drug delivery and biomedical engineering, due to their low cytotoxicity and higher uptake by targeted cells owing to their high affinity and specificity towards cell surface receptors. Despite the promising implications of this rapidly expanding field, there is no dedicated resource to study peptide nanostructures.ResultThis study endeavours to create a dedicated repository of short peptides, which may prove to be the best models to study ordered nanostructures formed by peptide self-assembly. SAPdb has a repertoire of 1,049 entries of experimentally validated nanostructures formed by the self-assembling of small peptides. It includes 701 entries are of dipeptides, 328 entries belong to tripeptides, and 20 entries of single amino acid with some conjugated partners. Each entry encompasses comprehensive information about the peptide such as chemical modifications in the peptide sequences, the type of nanostructure formed, and experimental conditions like pH, temperature, and solvent required for the self-assembly of the peptide, etc. Further, our analysis has shown that the occurrence of aromatic amino acids favours the formation of self-assembling nanostructures, as indicated by a large number of entries in SAPdb contain aromatics amino acids. Besides, we have observed that these peptides form different nanostructures under different experimental conditions. SAPdb provides this comprehensive information in a hassle-free tabulated manner at a glance. User-friendly browsing, searching, and analysis modules are integrated for easy retrieval and comparison of data and examination of properties. We anticipate SAPdb to be a valuable repository for researchers engaged in the burgeoning arena of nanobiotechnology.AvailabilityThe database can be accessed on the web at https://webs.iiitd.edu.in/raghava/sapdb.

RareLSD is a manually curated database of lysosomal enzymes associated with rare diseases that maintains comprehensive information of 63 unique lysosomal enzymes and 93 associated disorders. Each entry provides a complete information on the disorder that includes the name of the disease, organ affected, age of onset, available drug, inheritance pattern, defected enzyme and single nucleotide polymorphism. To facilitate users in designing drugs against these diseases, we predicted and maintained structures of lysosomal enzymes. Our information portal also contains information on biochemical assays against disease-associated enzymes obtained from PubChem. Each lysosomal entry is supported by information that includes disorders, inheritance pattern, drugs, family members, active inhibitors, etc. Eventually, a user-friendly web interface has been developed to facilitate the users in searching and browsing data in RareLSD with a wide range of options. RareLSD is integrated with sequence similarity search tools (e.g. BLAST and Smith-Waterman algorithm) for analysis. It is built on responsive templates that are compatible with most of browsers and screens including smartphones and gadgets (mobile, iPhone, iPad, tablets, etc.).

MotivationIn last three decades, a wide range of protein descriptors/features have been discovered to annotate a protein with high precision. A wide range of features have been integrated in numerous software packages (e.g., PROFEAT, PyBioMed, iFeature, protr, Rcpi, propy) to predict function of a protein. These features are not suitable to predict function of a protein at residue level such as prediction of ligand binding residues, DNA interacting residues, post translational modification etc.ResultsIn order to facilitate scientific community, we have developed a software package that computes more than 50,000 features, important for predicting function of a protein and its residues. It has five major modules for computing; composition-based features, binary profiles, evolutionary information, structure-based features and patterns. The composition-based module allows user to compute; i) simple compositions like amino acid, dipeptide, tripeptide; ii) Properties based compositions; iii) Repeats and distribution of amino acids; iv) Shannon entropy to measure the low complexity regions; iv) Miscellaneous compositions like pseudo amino acid, autocorrelation, conjoint triad, quasi-sequence order. Binary profile of amino acid sequences provides complete information including order of residues or type of residues; specifically, suitable to predict function of a protein at residue level. Pfeature allows one to compute evolutionary information-based features in form of PSSM profile generated using PSIBLAST. Structure based module allows computing structure-based features, specifically suitable to annotate chemically modified peptides/proteins. Pfeature also allows generating overlapping patterns and feature from whole protein or its parts (e.g., N-terminal, C-terminal). In summary, Pfeature comprises of almost all features used till now, for predicting function of a protein/peptide including its residues.AvailabilityIt is available in form of a web server, named as Pfeature (https://webs.iiitd.edu.in/raghava/pfeature/), as well as python library and standalone package (https://github.com/raghavagps/Pfeature) suitable for Windows, Ubuntu, Fedora, MacOS and Centos based operating system.

Insect neuropeptides and their associated receptors have been one of the potential targets for the pest control. The present study describes in silico models developed using natural and modified insect neuropeptides for predicting and designing new neuropeptides. Amino acid composition analysis revealed the preference of residues C, D, E, F, G, N, S, and Y in insect neuropeptides The positional residue preference analysis show that in natural neuropeptides residues like A, N, F, D, P, S, and I are preferred at N terminus and residues like L, R, P, F, N, and G are preferred at C terminus. Prediction models were developed using input features like amino acid and dipeptide composition, binary profiles and implementing different machine learning techniques. Dipeptide composition based SVM model performed best among all the models. In case of NeuroPIpred_DS1, model achieved an accuracy of 86.50% accuracy and 0.73 MCC on training dataset and 83.71% accuracy and 0.67 MCC on validation dataset whereas in case of NeuroPIpred_DS2, model achieved 97.47% accuracy and 0.95 MCC on training dataset and 97.93% accuracy and 0.96 MCC on validation dataset. In order to assist researchers, we created standalone and user friendly web server NeuroPIpred, available at (https://webs.iiitd.edu.in/raghava/neuropipred.)

BACKGROUNDIn the past, large numbers of methods have been developed for predicting secondary structure of proteins. Best of author’s knowledge no method has been specifically developed for predicting secondary structure of peptides. We analyzed secondary structure of peptides and proteins; it was observed that same peptide in protein adopt different secondary structures. Considering the wide application of peptides in therapeutic market, we made attempt to develop a method called PEP2D for predicting secondary structure of peptides.RESULTSIn this study, 3107 unique peptides have been used to train, test and evaluate peptide secondary structure prediction models. It was observed that regular secondary structure content (e.g., helix, beta-sheet) increased with length of peptides. Firstly, models based on various machine-learning techniques have been developed using binary profile of peptides and achieved maximum overall accuracy (Q3) 79.5%. The performance of models further improved from 79.5% to 83.5% using evolutionary information in the form of PSSM profile. We also evaluate performance of protein secondary structure prediction method PSIPRED on our dataset and achieved maximum accuracy 76.9%; particularly poor (Q3 71.4%) for small peptides having length less than 10 residues. Overall, PEP2D has better prediction of beta-sheets (Q3 74%) and coil region (Q3 87%) of peptides as compare to PSIPRED (Q3 54.4% for beta-sheet and Q3 77.9% for coil). We also measure performance of PSIPRED and PEP2D in terms of segment overlap (SOV); achieved 69.3 and 76.7 respectively.CONCLUSIONOur observations indicate that there is a need of developing separate method for predicting secondary structure of peptides. It was also observed that models based on PSSM profile perform poor on small peptides in comparison to long peptides. Based on our study, we developed method for predicting secondary structure of peptides. In order to provide service to user, a webserver/standalone has been developed (https://webs.iiitd.edu.in/raghava/pep2d/).