Last modified: February 17 2017.

hpchelp@iitd.ac.in

Research: Active user groups

    Click here to expand all details. Please send your inputs to zmanish@cc.iitd.ac.in
  1. Computer Services Centre
    Dr. Manish Agarwal:
    1. Classical Molecular Dynamics:
      • Simulations of substituted polyethylene using LAMMPS in collaboration with Gourav Shrivastav (Late Prof. Chakravarty's Lab, Department of Chemistry).
      • Simulations of Double Pervoskite materials in collaboration with Uzma Anjum (Dr. Ali Haider's lab, Department of Chemical Engineering)
      • Simulations of Amino acids near metal surfaces in collboration with Madhulika Gupta (Late Prof. Chakravarty's Lab, Department of Chemistry) and Shelaka Gupta (Dr. Haider Ali's Lab, Department of Chemical Engineering)
    2. Improving analysis of trajectories of Molecular Dynamics Simulations using GPU accelerated computing. In collaboration with Gourav Shrivastav of Department of Chemistry (Late Prof. Chakravarty's Lab) , we use CUDA to accelerate calculation of typical dynamical and static properties from molecular dynamics trajectories
    3. Improving acceleration of Direct Numerical Simulation programs in collaboration with Nishant Prashar from Dr. Sawan Suman's lab (Department of Applied Mechanics)
  2. Kusuma School of Biological Sciences
    1. Dr. Manidipa Banerjee's Group

      Dr. Manidipa Banerjee's group is using both experimental and computational approaches to understand the mechanisms and dynamicity of various processes of viruses, including virus assembly and disassembly, and virus interaction with host-cell receptors and membranes. We intend to use these studies to design virus-based macromolecular complexes for various biomedical applications like vaccines and drug delivery systems.

      We have been utilizing the HPC cluster at IIT for our computational work, which falls under two broad categories:

      • Molecular dynamics simulation studies to understand how viral peptides interact with biological membranes at the atomic scale: Simulation software packages like GROMACS and AMBER are being used for this purpose.
      • Cryo-electron microscopy-single particle reconstruction for near-atomic structure determination of virus-like particles: For this we have been using the image processing suites EMAN2, RELION and Scipion.

  3. Department of Physics
    Dr. Saswata Bhattacharya
    Research Application Area: Clusters and catalysis, hydrogen storage, electronic, optical, mechanical and vibrational properties of graphene like 2D materials, Thermoelectric materials, etc.
    • Dr. Saswata Bhattacharya’s group works in inter-disciplinary areas of condensed matter physics with broad research interest in first principles based simulation of designing new materials and understanding their properties using state-of-the-art density functional theory (DFT) (and beyond methods) combined with quantum chemistry approaches (eg. MP2, CCSD, etc.). In order to capture the effect at finite temperature and pressure ab initio atomistic thermodynamics and beyond approaches are used. Method developments: Genetic algorithm, Machine learning, Capturing anharmonic effects in free energy contributions.

  4. Department of Chemistry
    1. Late Prof. C. Chakravarty's Group
      Molecular Simulations of Liquids: Water-like fluids, Aqueous and Non-aqueous Solutions, Biomolecular Hydration, Nanoparticle Self-Assembly
      • Water and Other Tetrahedral Liquids: Debdas Dhabal
        Study triplet correlation functions in water.
      • Water and Aqueous Solutions: Saurav Prasad
        Crystallization and vitrification of water and aqueous solutions.
      • Nanoparticle Solvation and Effective Pair Interactions: Hari O. S. Yadav
        Self-assembly, under equilibrium or non-equilibrium conditions.
      • Nanoparticle Assembly in Organic Solvents: Gourav Srivastav
        Structural, transport and thermodynamic properties of organic solvents.
        NVIDIA Case study: GPU Usage in Molecular Dynamics Trajectory Analysis
      • Conformational Free Energy Landscapes: Divya Nayar
        Rigid-body water models in the neighbourhood of a folded peptide.
      • Biomolecular Hydration: Madhulika Gupta
        Simulations of Trp-cage miniprotein in different water models.
    2. Dr. Hemant Kashyap's Group:
      Molecular Simulations using Gromacs and NAMD
    3. Dr. Shashank Deep's Group
      Molecular Simulations:
      Investigation of the mechanism of protein stabilization and aggregation.
      • (Nidhi Katyal and Nidhi Kaur Bhatia)

        The malfunctioning of proteins is a root cause of various diseases. It is widely accepted that this behaviour is intimately tied to protein instability and aggregation. Thus, understanding the mechanism by which known external agents (eg. Trehalose) stabilizes the native state of protein and/or decelerates aggregation is of paramount importance in the design of new therapeutic stabilizers. In this perspective, our group examines the behaviour of different model proteins/peptides in the absence and presence of additives under broad temperature and concentration regime with the help of molecular dynamics simulations with Gromacs software available on HPCA.

        Recently, we have published an article entitled “Revisiting the conundrum of trehalose stabilization” in PCCP with simulations partially carried out on the HPCA cluster. The system consisted of 50000-70000 atoms and utilized 1 node, 2 GPUs and 16 processes with the performance of 13ns/12hrs. In this study, we have focused on the effect of trehalose against thermal denaturation on Lysozyme. We have monitored the structural properties of protein both in the absence and presence of trehalose under different temperature and hence constructed the free energy landscape using property space approach. An attempt towards dissecting the parameters crucial in orchestrating the stabilization process was carried out using principal component analysis. The clustering tendency of trehalose and its entrapment around lysozyme was viewed at a deeper level under different concentration ranges. We have also looked at effect of trehalose on water by calculating properties like tetrahedral order parameter, diffusion coefficients, radial distribution functions etc. All the calculations were done either using in-house codes or using gromacs analysis commands. Currently, we are trying understand the mechanism of protein self assembly.

    4. Prof. B. Jayaram
      (Debarati Dasgupta) A Generic View of the Factors Favoring Protein Folding.
  5. Department of Chemical Engineering
    1. Dr. Gaurav Goel's Group:
      Molecular Simulations using Gromacs
    2. Dr. Haider Ali's Group
      Calculations using Materials Studio and VASP
      • Integrated Fermentation and Catalysis
        (Shelaka Gupta, Dr. Tuhin Khan, Dr. Md Imteyaz Alam) Density functional theory (DFT) calculations to explain the catalytic ring-opening and decarboxylation of fermentation derived 2-pyrones and catalytic hydrogenation of lignin derived phenol.
        Shelaka Gupta; Md. Imteyaz Alam, Tuhin Suvra Khan; Nishant Sinha, M. Ali Haider, "On the mechanism of retro-Diels-Alder reaction of partially saturated 2-pyrones to produce biorenewable chemicals", RSC Advances, 6, 60433, 2016
        Shelaka Gupta, Rishabh Arora, Nishant Sinha, Md. Imteyaz Alam and M. Ali Haider, "Mechanistic insights into the ring-opening of biomass derived lactones”, RSC Advances, 6, 12932, 2016
      • Development of SOFC Electrodes
        (Uzma Anjum, Dr. Tuhin Khan, Dr. Manish Agarwal) Combined MD and DFT Simulations to understand the mechanism of oxygen reduction and migration in anisotropic layered perovskite structure electrodes for Solid Oxide Fuel Cells (SOFCs).
        Uzma Anjum, Saumye Vashishtha, Manish Agarwal, Pankaj Tiwari, Nishant Sinha, Ankit Agarwal, S. Basu, M. Ali Haider, “Oxygen anion diffusion in double perovskite GdBaCo2O5+δ and LnBa0.5Sr0.5Co2-xFexO5+δ (Ln=Gd, Pr, Nd) electrodes”, International Journal of Hydrogen Energy, 41, 7631-7640, 2016
        Uzma Anjum, Saumye Vashishtha, Nishant Sinha, M. Ali.Haider, "Role of Oxygen Anion Diffusion in Improved Electrochemical Performance of Layered Perovskite LnBa1-ySryCo2-xFexO5+δ (Ln = Pr, Nd, Gd) Electrodes" Solid State Ionics 280, 24–29, 2015
      • Carbon Dioxide Reduction to Produce Methanol (Neetu Kumari) Ceria based materials have been suggested to play an active role in the catalytic and electrocatalytic reduction of CO2. Role of CeO2(110) surface for the reduction of CO2 to CO and methanol is studied by DFT calculations.

        N. Kumari, Nishant Sinha, M. Ali Haider, S. Basu, “CO2 Reduction to Methanol on CeO2 (110) Surface: a Density Functional Theory Study", Electrochimica Acta, 177, 21–29, 2015.
        Neetu Kumari; M. Ali Haider; Manish Agarwal; Nishant Sinha, Suddhasatwa Basu, "Role of Reduced CeO2(110) Surface for CO2 Reduction to CO and Methanol", Journal of Physical Chemistry C, 2016
      • Natural Gas conversion to Aromatics

        (Sonit Balyan, Sourabh Mishra, Dr. Tuhin Khan) DFT calculations to understand the functioning of zeolite supported MoxCy nanocluster for non-oxidative dehydroaromatization of methane.
      • Biogenic Deactivation of Heterogeneous Catalysts

        (Madhulika Gupta, Fatima Jalid, Shelaka Gupta, Dr. Tuhin Khan, Dr. Manish Agarwal) All-atom molecular dynamic simulations to study the non-bonded interactions of amino acids with the metal catalyst surface.
      • Graphite Exfoliation to Produce Graphene

        (Afkham Mir, Dr. Tuhin Khan) DFT study of monolayer graphene exfoliation from graphite in solvents.
      • Development of Bio-Fuels

        (Gourav Shrivastav, Dr. Tuhin Khan, Dr. Manish Agarwal) MD simulations to calculate viscosity, isothermal compressibility and miscibility of biomass derived alkyl levulinates to be used as a fuel additive.
        Ejaz Ahmad, Md. Imteyaz Alam, K.K. Pant and M. Ali Haider, "Catalytic and Mechanistic Insights into the Production of Ethyl Levulinate from Biorenewable Feedstocks", Green Chemistry, 2016



    3. Dr. Sudip Pattanayek's Group
      Uses computational tools for three different areas, Polymer nano-composite, adsorption of protein and rheology of shear thickening fluid.
      • Polymer nanocomposite(Sunil Kumar): MD simulation with DREIDGING force field parameters are used for this problem. We have been simulating various polymers such as polyethylene, polypropylene, polyether-ether ketone etc as dispersion media and nano particles of various shape such cylindrical (such as Carbon Nano Tube, CNT), plate like (such as graphene sheet), spherical (such as bulky ball). We intend to see the organization of the polymers near the nanoparticles. Subsequently we observe the effect of nano particles on the mechanical, barrier properties of the nanocomposite. Polymer nanocomposites with polypropylene are also being studied in the group (Odath Sukumaran Ravishankar). A typical set of simulations of Polyethylene in Single Wall CNT(SWCNT) are shown below.
      • Adsorption of protein(Ronika Goswami): Understanding of of adsorption of protein on surfaces plays a major role in various biological applications such as biocompatibility of biomaterials, binding of proteins to to detect various biological events etc. We are using Self Consistent Field calculation and Molecular Dynanmics simulations to see the adsorption behavior of proteins on various substrates
      • Rheology of shear thickening fluid(STF)(Toshi Gupta Maheshwari): Silica particles in Poly Ethylene Glycol leads to shear thickening behavior at high shear rate. We have been using LAMMPS MD simulation package and Brownian dynamics simulation to optimize the particle size leading to STF behavior at high shear rate.
    4. Dr. Vivek Buwa's Group:
      Flow simulations using OpenFOAM and Ansys Fluent.
      (Arpit Jindal) Simulating liquid distribution in Trickle bed using commercial solver Fluent and verifying the results with corresponding experiments.
  6. Department of Applied Mechanics
    1. Dr. Sawan Suman's group:
      Direct numercial simulations.
      • Direct numerical simulation using COMPACT finite difference method scheme using MATLAB.(Sagar Saroha with Dr. Sawan Suman and Dr. Balaji Srinivasan)
        The Navier-Stokes equation is solved for Turbulent flows using numerical techniques like Finite Difference Method to capture physics of the fluid motion. Presently the computational code is serial; in future the Parallel Code utilizing multi-core and GPUs shall be run on cluster. The software will get extended to FORTRAN, PYTHON.
      • Direct Numerical Simulation of Navier-Stokes Equations. (Mohammad Danish with Dr. Sawan Suman and Dr. Balaji Srinivasan)
        We perform direct numerical simulation (DNS) for Navier-Stokes equations using gas kinetic method based solver. The computational domain is a cube of size 2π with 2563 cells and periodic boundary conditions imposed on the opposite sides. We use the simulated results to study the evolutions of various quantities of interest in decaying isotropic turbulence.
    2. Prof. Anupam Dewan Group:
      (Pritanshu Ranjan) Use of Variable Resolution turbulence modelling to study heat transfer over rectangular prisms which involves all complex flow phenomenon, for e.g, stagnation, local acceleration, separation, reattachment and vortex shedding. Use of Large Eddy Simulation to study the buoyancy driven flow.
  7. Centre for Industrial Tribilogy
    Dr. R. K. Pandey(Department of Mechanical Engineering) and Prof. V. K. Agarwal
    Elastohydrodynamic Lubrication of Concentrated contacts e.g. ball bearings using an in-house code.(Vivek Bharadwaj)
  8. Department of Mechanical Engineering
    1. Dr. B. Premachandran's group
      An in-house code has been developed to investigate boiling flow problems with sharp interfaces.
      • (BM Ningegowda) For capturing sharp interface, a Combined Level Set and Volume of Fluid method (CLSVOF) has been developed. The solver has been validated various standard test cases for two-phase flow without and with phase change. More detailed discussion on this solver has been presented here
    2. Dr. Devendra K. Dubey
      Molecular Simulations using NAMD
      • (Hariprasadh Vr) Dr. Dubey’s group uses large scale molecular dynamics simulations for understanding mechanics of bone and bone related disorders at nanoscopic lengthscales. One of the study investigates mechanisms and chemo-mechanics of fragility disorder in bone (Osteogenesis Imperfecta) which is associated with mutations in collagen (Col) and changes in the morphology of hydroxyapaptite (HAP) mineral compared to the normal bone. This requires self-assembly behavior study of Col-HAP systems using molecular dynamics simulations in different configurations. Figure shows a representative image of the constituents involved in atomistic Col-HAP supercell simulations.
  9. Department of Computer Science
    VisualSFM
  10. Department of Electrical Engineering
    1. Finding consistent patterns of information flow in Human Brain using fMRI
      Dushyant Sahoo with Dr. Rahul Garg(CSE) Functional Magnetic Resonance Imaging (fMRI) is being used to find functional connectivity and information flow in human brain. We aim to find all the networks of information flow when a person is at rest. In order to find consistent networks we are using granger causality analysis and hypothesis testing. The analysis is done on 500 subjects from 1000 functional connectome project.
    2. Convolutional Neural Networks for Image Classification
      • (Saurav Gupta with Prof. Santanu Chaudhury) Convolutional Neural Networks(CNNs) are a type of feed-forward artifcial neural networks with a tiled architecture with sparse connections and some tricks like shared weights to reduce number of parameters and learn useful features. Then, max pooling and subsampling layers are introduced, so as to achieve translational invariance and reduce the amount of computation. This project uses convolutional neural networks for image classifcation with the help of Theano and Pylearn2. Theano is a CPU and GPU expression compiler with a Python wrapper and Pylearn2 a machine learning library based on Theano.
    3. Dr. Manik Verma (Department of Computer Science and Engineering)
      Object detection and classification in images using Convolutional Neural Network (Rahul Kumar)
      Softwares Used: Intel MKL, Matlab CPU or CUDA 5.5/6.0
      Experimenting with Convolution Neural Network on ImageNet dataset to detect and classify objects in images. Details of ImageNet dataset are available here.
  11. Department of Civil Engineering
    1. Dr. Dhanya C. T.'s group
      Regional scale hydrologic modelling for climate change impact assessment using Variable Infiltration Capacity (VIC) Macroscale Hydrologic Model
  12. Department of Biochemical Engineering and Biotechnology
    1. Dr. D. Sundar's group
      Long term goal of Dr. D Sundar's group is to understand DNA-protein interactions to evolve DNA-binding specificity in proteins and to make highly specific custom-designed proteins for various applications in Genome Engineering. The current areas of focus include: (a) rational design of zinc finger proteins (ZFPs) for genome editing (b) Biological activity of natural drugs (c) Metabolic Engineering. We use both computational and wet-lab experimental approaches in close collaboration with several groups in India and abroad.