Prescience provides computational prediction of material properties with high accuracy even before they are synthesized. We capitalize on the advances in machine learning (ML) for computational studies of materials. We provide services to the soft matters compromising any of these materials: Surfactants (ionic/non-ionic), Proteins and Lipids (charged/neutral), Polymers, and ionic liquids.
Deep learning has been receiving increasing attention and has achieved great improvements in both time efficiency and prediction accuracy. It is well known that computational simulation and experimental measurement are two conventional methods that are widely adopted in the field of materials science. However, it is difficult to use these two methods to accelerate materials discovery. Prescience is acting as a bridge for computational results, machine learning and data mining or by close collaboration between computational predictions and experimental validation. The future vision is to tweak our deep learning models as required to match the highly versatile need of the industry.
In heterogeneous catalysis the phase of the catalyst differs from that of the reactant. Heterogeneous catalysis is widely used in these four sectors of chemical industries (1) Polymer industry (2) Coal, oil and gas refining (3) Chemical manufacturing industry (4) Environmental application. Most of the heterogeneous catalyst are in solid phase and reacts with gas phase reactants. Such instances involve cycles of molecular adsorption, reaction and desorption which occurs on the molecular surface. Hence thermodynamic and kinetic aspects of this absorption, desorption process is an import factor in understanding the efficiency of heterogeneous catalysts. Here comes the importance of computational methods for designing and understanding the activity of heterogeneous catalyst which are being widely used in chemical industries. Prescience Insilico Pvt. Ltd. can provide valuable insight in understanding the thermodynamics, mass transfer and heat transfer influencing the rate of these catalytic reaction in large scale industrial processes. Understanding the intricate mechanistic aspects of catalysis on metallic (transition metal) and non-metallic surfaces is our another important expertise. Both by using quantum chemical methods and molecular dynamics we can provide valuable insights on the thermodynamic and kinetic aspects of absorption processes of polymers and other gas phase reactants on heterogeneous catalyst surface. Besides theoretical prediction of detailed reaction mechanism involving double bond hydrogenation, C-H bond activation in chemical processes catalyzed by heterogeneous catalysts is another important service provided by us.
A computer simulation methods are very useful to predict the relative change in properties of a base fuel and lubricants in the presence of additives (with varying concentration and physical conditions, e.g. temperature and pressure). Particle based multiscale simulations methods could be used to identify (in addition to change in properties) new chemical structures, configuration, composition and topologies of the additives e.g., polymers which can be added to the fuel for the desired modification of the viscosity to meet specific requirements.
The multiscale method could be all atomistic molecular simulations, coarse grained and dissipative particle dynamics (DPD) along with reptation dynamics. Several analysis methods e.g., Green Kubo formalism on the production run trajectories of the equilibrated system could be used to calculate the average rheological properties.
The method is fast (because of mesoscale length and time scales) and predicts the properties as function of chemistry, topology, concentration, temperature, pressure and/or a combination of these. This allows for a screening for new chemistries that can be synthesised and provide a cost effective additives fuel and lubricants.
The solar energy shining on a solar cell or a photovoltaic (PV) cell is converted into the usable electricity. Improving this conversion efficiency is a key goal of research and helps make PV technologies cost-competitive with conventional sources of energy.
Finding a catalyst that can provide desired product in good selectivity and yield is often the beginning of a long run in many chemical industries. Homogeneous catalysts are soluble in reaction medium. In all chemical industries ranging from refineries to pharmaceuticals, from fossil fuels to biomass over 90% of all chemical products have at least one catalytic step in their manufacture. With increasing need of more sustainable and environmental friendly way of developing catalysts, computational methods for modelling homogenous catalysts with optimal activity have become immensely popular these days. Both quantum chemical methods and molecular dynamic techniques can be used to effectively shed light in designing homogeneous catalyst in the area of biomass utilisation, pharmaceutical intermediates, manufacturing of organic chemicals and fine chemicals like, flavours and fragrances. Prescience Insilico Pvt. Ltd. has the capacity to sustainably design homogeneous catalysts for processes that span high temperature, high-pressure, high-tonnage to small-scale ambient-pressure liquid phase at reasonable cost. Prescience can also provide meaningful insights about the mechanisms of these catalytic processes right from the atomic level, with quantum chemistry tools, with an aim of contributing to (1) human healthcare, (2) green & sustainable chemistry and (3) catalytic chemistry. We can see the breadth and depth of the catalysis development needed in the modern chemical industries and can deliver new ideas to optimize the systems for conclusive experiments that reduce generation of chemical waste and saves manpower, energy, and money.
Drug discovery is an extremely long, expensive and challenging process that takes an average of 10 to 15 years with a very low success rate ranging from 2% to 10%. With an increase in the knowledge about the molecular mechanisms behind different diseases/disorders, more potential targets for the design of new drugs are being identified. Our drug discovery pipeline aims to accelerate the development process that includes identifying new chemical entities (NCE) or rapid repurposing of drugs for direct therapeutic approaches. This approach takes advantage of existing information for the approved small molecules and biologics, and enables rapid clinical trials or regulatory review. The unique in-house developed Multi-Target Multi Ligand platform can speed up drug development and generate potential new drug compounds. With our proprietary platform, we identified multiple drug molecules with high potential that target viral and host proteins for SARS-CoV2, including a list of new chemical entities and FDA approved drugs. In combination with data mining and machine learning platform, our software can generate potential drug candidates for several other diseases
BioChem-informatics software is an Artificial Intelligence (AI) powered drug molecule predictor for target proteins inhibited by drug molecule. Generation of a target-specific drug molecule dataset and understanding their probability of inhibiting that target is an essential element in modern biomedical research. This software is an amalgamation of two different approaches: Bioinformatics and Cheminformatics. We incorporate protein sequence based automated generation of target-specific drug datasets either from open-source curated databases of bioactive molecules with drug-like properties or by using machine-learning and deep-learning models. An additional module in this software enables protein structure refinement capabilities, i.e., automated loop modelling, generate missing residues in PDB files, hydrogen addition, and loop refinement.
Our unique computational modelling method quantifies the physicochemical properties of chemical compounds to prioritize a large number of molecules for virtual screening protocol. This software overcomes the limitation of high computational cost required for quantum chemical methods, and generates systemic and cost-effective solutions through in-house developed data mining and machine learning approach.
With the ever-growing knowledge about the human genome data, pharmaceutical companies face challenging tasks in handling the copious data generated in parallel. Prescience provides graphical and web-based programs for analyzing biological data such as genetic sequence similarity, protein data mining and drug discovery. Our in-house computational biology platform has integrated diverse modules such as MTMLESS, Bio-Chem Informatics to analyze large scale biological and chemical data sets for accelerated drug discovery.