Crystal Structure Prediction

In silico polymorph discovery of novel molecules using Al/ML, GPU acceleration & DevOps powered automation on cloud infrastructure
Structural representation of Rotigotine (C19H25NOS)
I: 2D representation
II: Stable conformers
III: Stable polymorphs
Context

Accelerated workflow for a free energy landscape scan

Identifying polymorphs with the desired physiochemical properties demand a comprehensive scan of the free energy landscape of an API molecule in its crystalline solid state.
Such an undertaking requires implementing computationally expensive Quantum mechanical/chemical methods. This expense is compounded multifold if the polymorph studies include salts, cocrystals, hydrates/solvates, etc.
Aganitha utilizes advanced AI/ML models in tandem with GPU-based QM software, all running on elastic/scalable cloud infrastructure, to aid in exploring the energetics of a range of candidate crystal structures within shorter time frames. These can serve as valuable computational aids in the formulation efforts of your product.
Our Solution

Scalable and accelerated in silico pipeline to identify stable polymorphs

Use our pipeline for a cost-effective in silico crystal structure screening. We combine the concepts of Quantum chemistry with advances in Al/ML, Cloud & DevOps. Key features of our solution include:
  • Generative models to navigate the conformational space of an organic molecule
  • Diffusion based models to generate candidate crystal structures
  • Graph Neural Networks (GNN) based models to predict Lattice Energy
  • Accelerated DFT computational pipelines to screen candidate crystal structures
Crystal Structure Prediction pipeline
Highlights

Key components of Aganitha’s Crystal Structure Prediction pipeline

Pipeline with comprehensive capabilities

Generation of diverse candidate crystal structures starting from multiple stable 3D conformers for a more comprehensive exploration of the crystal structure landscape

State of the art AI/ML tools & techniques

Diffusion models to generate candidate crystal structures and GNN based for Lattice Energy prediction models

System specific customization 

Modules built to leverage open-source packages, AI/ML models, GPU based QM packages

Outcomes

Swift and data safe polymorph discovery

Fast and Cost-effective

Diffusion models & GNN models drive identification of polymorphs thereby rapidly getting you to your end result.

Data Privacy and safety

We bring infrastructure as code to your data in your environment ensuring that your data is safe

Configurable and Scalable 

Scalable computational resources with on-demand cloud-based High Performance Computing (HPC) clusters workload management techniques

Download our case study on stable conformer identification

Discover our offerings across the biopharma value chain

Learn more about Crystal Structure Prediction