Crystal Structure Prediction

Crystal Structure Prediction (CSP)

De-risk your solid form selection process by identifying the most stable polymorph at room temperature

Stability ranking of crystal polymorphs

Overcome the risks associated with disappearing polymorphs in late stage drug development. Schrödinger’s proprietary crystal structure prediction (CSP) platform identifies the stable crystal polymorphs at 0K and RT for a given active pharmaceutical ingredient (API). The CSP platform can now be used to predict crystal polymorphs in the Z’=2 search space, salts and monohydrate polymorph systems.

Key Capabilities

Fast and comprehensive identification of polymorphs at room temperature and beyond

Novel, systematic approach allows exhaustive yet efficient sampling of crystal packings
High throughput workflow with quick turnaround time to support CADD and CMC teams working on lead optimization (CSP for scaffold design), polymorph screening experiments (CSP-DoE), and manufacturing processes (CSP-derisk)
Includes advanced workflows for the prediction of anhydrous Z’=2, monohydrate, solvate, and salt forms of the API

High accuracy validated on extensive dataset of challenging, diverse drug-like molecules

Retrospective validation on a set of 66 drug-like molecules with 137 experimental polymorphs and an accuracy close to 100% in predicting the most stable solid form
Prospective validation on a central nervous system drug molecule showed high accuracy and reliability

Crystal Structure Prediction Workflow

Schrödinger solutions for physicochemical property prediction

Optionally predict key properties of an API to support selection of a stable solid form.

Crystalline solubility of polymorphs using free energy methods with FEP+

Learn more

ssNMR chemical shifts to support crystallization experiments with Quantum ESPRESSO

Learn more

Crystal habits (morphology) to support downstream processing with MS Morph

Learn more

Mechanical properties (Young’s and shear modulus) to complement direct compaction and milling experiments with MD engine Desmond

Learn more

Featured Service

Crystal Structure Prediction Services

Work with our team of computational experts to de-risk your solid form selection process. Starting from a 2D structure of the API, Schrödinger’s team will deliver to you the thermodynamic stability ranking of crystal polymorphs.

Speak to an expert

Featured Webinar

Life Science
Webinar

Crystal structure prediction workflow for small molecule drug formulation

Watch webinar

Publications

Browse the list of peer-reviewed publications using Schrödinger technology in related application areas.

A robust crystal structure prediction method to support small molecule drug development with large scale validation and blind study

Zhou D, et al. Nature Communications, 2025, 16, 2210

Learn more

Free energy perturbation approach for accurate crystalline aqueous solubility predictions

Hong RS, et al. J. Med. Chem. 2023, 66, 23, 15883-15893

Learn more

Novel physics-based ensemble modeling approach that utilizes 3D molecular conformation and packing to access aqueous thermodynamic solubility: A case study of orally available bromodomain and extraterminal domain inhibitor lead optimization series

Hong RS, et al. J. Chem. Inf. Model. 2021, 61, 3, 1412-1426

Learn more

Software and services to meet your organizational needs

Software Platform

Deploy digital drug discovery workflows using a comprehensive and user-friendly platform for molecular modeling, design, and collaboration.

Learn more

Modeling Services

Leverage Schrödinger’s computational expertise and technology at scale to advance your projects through key stages in the drug discovery process.

View services

Support & Training

Access expert support, educational materials, and training resources designed for both novice and experienced users.

Explore Resources

Target Enablement Services

Speak to an expert View all services

Unlock your protein target and ligand series of interest for FEP+

Free up your time and resources — let us prepare your protein-ligand system of interest for structure-based design. Whether you are starting from experimental X-ray structures, cryo-EM structures, AlphaFold predicted structures, or homology models, Schrödinger scientists will leverage our unique technologies and expertise to optimize and validate protein-ligand complexes for highly predictive structure-based design.

Best suited for companies and teams:

• Who want to accelerate hit-to-lead and lead optimization to achieve program milestones
• Who want to minimize time spent on preparing and validating structural models
• Who want to leverage Schrödinger’s advanced physics-based methods and expertise

Best suited for projects:

• Where SAR data is available and location of the binding site is known, but structural data of the target is incomplete (cryo-EM, low resolution X-ray) or missing (homology model, AlphaFold model)
• Where the chemistry is evolving away from the original project hits or lead series and structural data is becoming suboptimal

Propel your discovery program with unrivaled technologies and expertise

Create design-ready structures for your unique on-target or off-target

Schrödinger has successfully enabled use of FEP+ for a wide range of target classes, including GPCR, kinases, and more

Benefit from the full impact of Schrödinger technologies at scale

Service includes all computing, licensing, and service hours to perform comprehensive protein target assessment and build accurate structural models

Complement your unique target knowledge with our computational expertise

Schrödinger brings our years of expertise in enabling challenging targets for structure-based drug design with FEP+

Case Study I

Challenge:

Despite a ~3.2Å cryo-EM structure of the target in complex with a ligand from the series of interest, known SAR could not be explained and the program was, effectively, not structurally enabled.

Result:

Our ensemble trial-and-error approach allowed us to identify an alternate ligand binding mode, as well as small but important protein adjustments necessary for binding. Although the binding mode was flipped compared to the model proposed by the structural biology team, it also fit the experimental density and, importantly, was predictive of all known SAR available.

Case Study II

Challenge:

A series of small molecule inhibitors against a flexible protein target was reaching its limits due to the lack of structural insights to guide the design, as well as off-target activity on a homologue.

Result:

Schrödinger enabled accurate FEP+ predictions for the series of interest both on the target and the anti-target. Importantly, it revealed specific loop and side-chain motions that were necessary for the ligands to bind, and remarkably, the results were counterintuitive as they revealed different binding modes in the two proteins despite their high similarity.

Enable FEP+ starting from all levels of structural information

Whether you are starting with template-based homology and AlphaFold models or if you have incomplete cryo-EM or experimental X-ray structures, Schrödinger’s team has the scientific expertise and computational resources needed to structurally enable your unique project.

Scientifically-validated solutions for structure-based drug discovery

Publication

Enabling structure-based drug discovery utilizing predicted models

Miller, E. B. et al. Cell. 2024, 187, 3, 521-525

Read paper

Publication

Using AlphaFold and experimental structures for the prediction of the structure and binding affinities of GPCR complexes via induced fit docking and free energy perturbation.

Coskun D, et al. J. Chem. Theory Comput. December 15, 2023.

Read paper

Publication

Benchmarking refined and unrefined AlphaFold2 structures for hit discovery.

Zhang Y, et al. J. Chem. Inf. Model. 2023, 63, 6, 1656–1667.

Read paper

Publication

Induced-fit docking enables accurate free energy perturbation calculations in homology models.

Xu T, et al. J. Chem. Theory Comput. 2022, 18, 9, 5710–5724.

Read paper

Enabling digital technologies to drive discovery programs

FEP+

High-performance free energy calculations for drug discovery

Life Science

IFD-MD

Accurate ligand binding mode prediction for novel chemical matter, for on-targets and off-targets

Life Science

Schedule a call to discuss Schrödinger’s Target Enablement Services

Partner with Schrödinger’s experts to ensure your project’s success.

Schedule a call to learn how our experts can help you access in silico solutions – let us help you prepare your protein target of interest for prospective FEP+ free energy calculations.

Form submitted

Thank you, we’ll be in touch soon.