Case Studies – SilicoScientia

Abstract

Artificial intelligence (AI) and machine learning (ML) are significantly
transforming drug discovery by allowing the exploration of an extra-large
chemical space, which speeds up and reduces risks in the traditionally costly
and time-consuming process. The SilicoXplore platform, developed by
SilicoScientia Private Limited, exemplifies this progress, providing a
complete, end-to-end drug discovery solution that incorporates over twenty
modules, many of which are specifically developed using ML. This platform
efficiently manages various drug discovery approaches, such as structure-based
and ligand-based drug design, pharmacokinetics, toxicity prediction, molecular
dynamics (MD) simulations, and quantum chemical calculations, for extensive
datasets. A key scientific insight that drives SilicoXplore’s effectiveness is
the integration of physics-based models with machine learning approaches,
combining mechanistic understanding with data-driven prediction power. Glioblastoma
multiforme (GBM), an aggressive and incurable brain tumor driven by EGFR,
indicates a poor prognosis due to its relentless nature. The SilicoXplore
platform was used to identify potential EGFR pathway modulators from existing
compound libraries. Out of 25000 from BindingDB, 19000 compounds were retained
after extensive curation. The ML-based SilicoScreen tool of SilicoXplore
produced a highly predictive, statistically robust model. The SimSearch tool
identified over 10000 active compounds similar to dataset actives, respecting the
non-redundancy of 55145 compounds. ToxAI and SilicoPhysChem pinpointed the top
30 molecules. Additionally, PharmFrag analyzed pharmacophoric features and
selected six promising molecules for EGFR-targeted glioblastoma therapy. Therefore,
SilicoXplore, which combines advanced ML methods with physics-based approaches,
efficiently explores large chemical spaces to find lead-like molecules without
the need for extensive computational resources or programming skills.

Myosin in Hypertrophic Cardiomyopathy

Abstract:
Hypertrophic cardiomyopathy (HCM) is an inherited dysregulation of specific β-cardiac
myosin protein functions that causes the sudden mis regulation of myosin heads and increases
the heart's contractility. It can lead to major cardiac failure and is seen in the majorly adult
generation. The mutations of the β-cardiac myosin protein are also responsible for HCM by
altering molecular interactions and increasing ATPase activity. So, for overcoming this
critical problem of HCM, in this research, we specifically targeted the R719W specific
mutation β-cardiac myosin protein for study, and we identified potent hit compounds as
R719W mutation β-cardiac myosin protein inhibitors using various Computational techniques
available in the SilicoXplore platform generated by the SilicoScientia team. From the
SilicoXplore platform, the modern techniques such as Protein modelling, Similarity search
using SilicoDatabase of natural compounds, Molecular docking using BINA docking, and
Toxicity prediction by Tox-AI tool of SilicoXplore are utilized for discovering novel small
molecules for the prevention and inhibition of myosin protein. The overall research is
smoothly performed and executed by the SilicoXplore Platform. The Molecular dynamics
Simulation and MMGBSA results also showed potent interaction between selected hit
compounds and the myosin protein as stable complexes. The final selected compounds are
potential and promising candidates for further innovation and experimental assessment in the
prevention of hypercontractility of the heart and HCM.

A Comprehensive Cloud-Based Platform for Drug Discovery, Case Study on EGFR-Targeted Glioblastoma Treatment

Structure-Based Discovery of Novel Inhibitors for the R719W Mutant β-Cardiac