A5078260902- Protein Structure and Dynamics
- Bioinformatics and Genomic Networks
- Enzyme Structure and Function
- RNA and protein synthesis mechanisms
- Genomics and Phylogenetic Studies
- Machine Learning in Bioinformatics
- Metabolism, Diabetes, and Cancer
- Cancer, Hypoxia, and Metabolism
- Protein Kinase Regulation and GTPase Signaling
- Microtubule and mitosis dynamics
- Biomedical Text Mining and Ontologies
- Cancer, Lipids, and Metabolism
- Natural Language Processing Techniques
- Topic Modeling
- Tuberculosis Research and Epidemiology
- Coenzyme Q10 studies and effects
- Advanced Proteomics Techniques and Applications
- Mitochondrial Function and Pathology
- Melanoma and MAPK Pathways
- Genetics, Bioinformatics, and Biomedical Research
- HER2/EGFR in Cancer Research
- Porphyrin Metabolism and Disorders
- Cancer Mechanisms and Therapy
- Gene Regulatory Network Analysis
- Phytochemistry and Bioactive Compounds
University of Michigan
2018-2020
Genpact (United States)
2020
University of Georgia
2011-2017
Excel Life Sciences (India)
2016
Indian Institute of Science Bangalore
2005-2013
Eukaryotic protein kinases (EPKs) regulate numerous signaling processes by phosphorylating targeted substrates through the highly conserved catalytic domain. Our previous computational studies proposed a model stating that properly assembled nonlinear motif termed Regulatory (R) spine is essential for activity of EPKs. Here we define required intramolecular interactions and biochemical properties R-spine newly identified "Shell" surrounds using site-directed mutagenesis various in vitro...
In recent years extracting relevant information from biomedical and clinical texts such as research articles, discharge summaries, or electronic health records have been a subject of many efforts shared challenges.Relation extraction is the process detecting classifying semantic relation among entities in given piece texts.Existing models for this task domain use either manually engineered features kernel methods to create feature vector.These are then fed classifier prediction correct...
We report that Tribbles 2 (TRB2), a pseudokinase linked to cancer, binds ATP and autophosphorylates in metal-independent manner. This activity is prevented by mutations the ATP-binding site TRB2 mutant can be targeted with small molecule ligand.
Abstract Background A polypeptide chain of a protein-protein complex is said to be obligatory if it bound another throughout its functional lifetime. Such might not adopt the native fold in unbound form. non-obligatory associates with and dissociates upon molecular stimulus. Although conformational changes at interaction interface are expected, overall 3-D structure unaltered. The present study focuses on complexes understand further differences between interfaces. Results known recognized...
Protein tyrosine kinases (PTKs) are a group of closely related enzymes that have evolutionarily diverged from serine/threonine (STKs) to regulate pathways associated with multi-cellularity. Evolutionary divergence PTKs STKs has occurred through accumulation mutations in the active site as well commonly conserved hydrophobic core. While functional significance variations is understood, relatively little known about how core contribute PTK evolutionary divergence. Here, using combination...
The protein kinase catalytic domain contains several conserved residues of unknown functions. Here, using a combination computational and experimental approaches, we show that the function some these is to maintain backbone geometry active site in strained conformation. Specifically, find catalytically important HRD motif deviates from ideality high-resolution structures results favorable hydrogen bonds with noncatalytic site. In particular, aspartate F-helix diverse eukaryotic...
The epidermal growth factor receptor (EGFR) is overexpressed in multiple carcinomas and the focus of a variety targeted therapies. Here we report design peptide-based compounds that mimic EGFR dimerization arm inhibit allosteric activation EGFR. These peptides are modified to contain triazolyl bridge between peptide strands constrain β-loop. In this study, demonstrate these have significantly improved proteolytic stability over non-modified sequence, their inhibitory effects dependent on...
Lack of large-scale efforts aimed at recognizing interactions between host and pathogens limits our understanding many diseases. We present a simple generally applicable bioinformatics approach for the analysis possible proteins parasite, Plasmodium falciparum, human host. In first step, physically compatible parasite are recognized using homology detection. This dataset putative in vitro is combined with datasets expression sub-cellular localization. integrated reduces drastically number...
A lack of information on protein–protein interactions at the host–pathogen interface is impeding understanding pathogenesis process. recently developed, homology search-based method to predict applied gastric pathogen, Helicobacter pylori between proteins H. and human proteinsin vitro. Many predicted could potentially occur pathogen its host during as we focused mainly pyloriproteins that have a transmembrane region or are encoded in pathogenic island those which known be secreted into host....
Mitogen activated protein kinases (MAPKs) form a closely related family of that control critical pathways associated with cell growth and survival. Although MAPKs have been extensively characterized at the biochemical, cellular, structural level, an integrated evolutionary understanding how differ from other is currently lacking. Here, we perform statistical sequence comparisons to identify features MAPK functional divergence. We show, for first time, virtually all MAPK-distinguishing...
Many prokaryotic kinases that phosphorylate small molecule substrates, such as antibiotics, lipids and sugars, are evolutionarily related to Eukaryotic Protein Kinases (EPKs). These Eukaryotic-Like (ELKs) share the same overall structural fold EPKs, but differ in their modes of regulation, substrate recognition specificity—the sequence determinants which poorly understood. To better understand basis for ELK specificity, we applied a Bayesian classification procedure designed identify...
Abstract We present a simple method for the analysis of large networks based on their graph spectral properties. One advantages this is that it uses single numerical computation to identify subclusters in connected graph, which can significantly simplify complexity involved analyzing graphs. This illustrated using network protein chains constructed basis structural similarities. The large‐scale properties and cluster subcluster organization chain are presented. summarize results functional...
Tumor initiation and progression are associated with numerous metabolic alterations. However, the biochemical drivers constraints that contribute to gene dysregulation unclear. Here, we present MetOncoFit, a computational model integrates 142 features can impact tumor fitness, including enzyme catalytic activity, pathway association, network topology, reaction flux. MetOncoFit uses genome-scale modeling machine-learning quantify relative importance of various in predicting cancer expression,...
Sensitive remote homology detection and accurate alignments especially in the midnight zone of sequence similarity are needed for better function annotation structural modeling proteins. An algorithm, AlignHUSH HMM-HMM alignment has been developed which is capable recognizing distantly related domain families The method uses information, form predicted secondary structure probabilities, hydrophobicity amino acids to align HMMs two sets aligned sequences. effect using adjoining column(s)...
Searches using position specific scoring matrices (PSSMs) have been commonly used in remote homology detection procedures such as PSI-BLAST and RPS-BLAST. A PSSM is generated typically one of the sequences a family reference sequence. In case searches sequence same query. Recently we shown that against database multiple family-profiles, with each members sequence, are more effective than classical single family-profiles. Despite relatively better overall performance when compared common...
PROtein Domain Organization and Comparison (PRODOC) comprises several programs that enable convenient comparison of proteins as a sequence domains. The in-built dataset currently consists approximately 698 000 from 192 organisms with complete genomic data, all the SWISSPROT obtained Pfam database. All entries in PRODOC are represented functional domains, assigned using hidden Markov models, instead amino acids. On average 69% proteomes 49% residues covered by domain assignments. Software...
In recent years extracting relevant information from biomedical and clinical texts such as research articles, discharge summaries, or electronic health records have been a subject of many efforts shared challenges. Relation extraction is the process detecting classifying semantic relation among entities in given piece texts. Existing models for this task domain use either manually engineered features kernel methods to create feature vector. These are then fed classifier prediction correct...