U
U. Deva Priyakumar
Researcher at International Institute of Information Technology, Hyderabad
Publications - 147
Citations - 2867
U. Deva Priyakumar is an academic researcher from International Institute of Information Technology, Hyderabad. The author has contributed to research in topics: Computer science & Nucleic acid. The author has an hindex of 24, co-authored 130 publications receiving 2111 citations. Previous affiliations of U. Deva Priyakumar include Pondicherry University & Indian Institute of Chemical Technology.
Papers
More filters
Journal ArticleDOI
Impact of 2'-hydroxyl sampling on the conformational properties of RNA: update of the CHARMM all-atom additive force field for RNA.
TL;DR: Application of the CHARMM36 model to a collection of canonical and noncanonical RNA molecules reveals overall improved agreement with a range of experimental observables as compared to CHARMM27, and indicates the sensitivity of the conformational heterogeneity of RNA to the orientation of the 2′‐hydroxyl moiety to support a model whereby the 2‐Hydroxyl can enhance the probability of conformational transitions in RNA.
Journal ArticleDOI
Computational Approaches for Investigating Base Flipping in Oligonucleotides
TL;DR: Over the years, numerous crystal structures of protein-DNA complexes where base flipping occurs have been reported, including several methyltransferases (M.HhaI,11,16 M.HaeIII,17 and M.TaqI18,19), glycosylases20,21 (T4 endonuclease V,22 human UDG,23-25 Escherichia coliMUG,26 human AAG,27 E. coli AlkA,28 and
Journal ArticleDOI
A computational study of cation–π interactions in polycyclic systems: exploring the dependence on the curvature and electronic factors
TL;DR: In this paper, Li+ and Na+ π-complexes of corannulene 2, sumanene 3CH2, heterosumanenes 3X, triphenylene 4 and heterotrindenes 5X have been analyzed with double and triple-ζ quality basis sets.
Posted ContentDOI
LigGPT: Molecular Generation using a Transformer-Decoder Model
TL;DR: The model, LigGPT, outperforms other previously proposed modern machine learning frameworks for molecular generation in terms of generating valid, unique and novel molecules and it is demonstrated that the model can be trained conditionally to optimize multiple properties of the generated molecules.
Journal ArticleDOI
Exploration of C6H6 Potential Energy Surface: A Computational Effort to Unravel the Relative Stabilities and Synthetic Feasibility of New Benzene Isomers†
TL;DR: In this paper, an exhaustive study on all these topological structures resulted in a total of 263 stationary points on the C6H6 potential energy surface, including 209 as minima, 31 as transition states, 8 as second-order, 7 as third-order and 1 as fourth-order saddle points.