US 11,783,920 B2
System and method for evaluation of at least one potential tastant from a plurality of tastants
Anukrati Goel, Pune (IN); Kishore Gajula, Pune (IN); Rakesh Gupta, Pune (IN); and Beena Rai, Pune (IN)
Assigned to TATA CONSULTANCY SERVICES LIMITED, Mumbai (IN)
Filed by Tata Consultancy Services Limited, Mumbai (IN)
Filed on Feb. 6, 2020, as Appl. No. 16/783,824.
Claims priority of application No. 201921004696 (IN), filed on Feb. 6, 2019.
Prior Publication US 2020/0251187 A1, Aug. 6, 2020
Int. Cl. G16C 20/10 (2019.01); G16C 20/30 (2019.01); G16C 20/64 (2019.01); G16C 20/50 (2019.01)
CPC G16C 20/10 (2019.02) [G16C 20/30 (2019.02); G16C 20/50 (2019.02); G16C 20/64 (2019.02)] 18 Claims
OG exemplary drawing
 
1. A processor implemented method of evaluating at least one potential tastant from a plurality of tastants, comprising:
receiving, via one or more hardware processors, information associated with a plurality of molecular activities, wherein the plurality of molecular activities corresponds to a taste index and a binding energy, wherein the binding energy correspond to one or more interactions between at least one molecule and at least one receptor;
generating, via the one or more hardware processors, a plurality of data-based models based on the known taste index associated with at least one tastant and information from associated molecular structure/descriptors, wherein the data-based models correspond to at least one of (i) structure property relationship models, (ii) machine learning models corresponding to a linear regression models or non-linear regression models;
classifying, via the one or more hardware processors, a new molecule based on the generated data-based models for the at least one tastant;
screening, via the one or more hardware processors, the one or more classified new molecules in an applicability domain of the generated data-based models based on physics-based models, developed at molecular scale, by at least one molecular modeling technique, wherein the molecular modelling techniques include at least one of a Monte Carlo simulation and molecular dynamics, wherein the molecular modeling technique screens the one or more classified new molecules using insights from the interaction of the at least one tastant with the at least one receptor;
evaluating, via the one or more hardware processors, the at least one potential tastant from the screened molecules based on at least one of a bioavailability and a toxicity; and
providing a framework to virtually screen the plurality of tastants from a plurality of databases and utilizes both the data-based models and the physics-based models for in-silico tastants discovery, design, classification, and testing the plurality of tastants,
wherein the framework, with a systematic approach, supports in speeding process of tastant design, development process.