Targeted Library
XtalPi's Targeted Library offerings integrate key structural features of therapeutically relevant molecules into synthetically feasible medicinal chemistry scaffolds, creating drug-like molecules based on real-world data. By combining novel scaffolds and building blocks, XtalPi expands chemical space and diversity, resulting in unique, synthesizable compounds. These libraries can be tailored for specific therapeutic areas, target classes, and small molecule modalities, such as noncovalent inhibitors, covalent inhibitors, ADCs, PROTACs, and molecular glues.
What we offer
Benefits of XtalPi Compound Libraries - Bespoke libraries to meet broad screening needs
-
Novelty and diversity
-
Enhanced chemical space
-
Rapid delivery - typically 600 compounds/week
-
Streamlined compound management for seamless logistics
-
In-house biology platform for rapid bioassay measurement
-
Robust quality - assured 10 mg scale with >90 % purity
-
Constantly expanding collection - new compound every 8 weeks
-
Enhanced Novelty & Diversity◦ Empowering scaffold hopping of reported biologically active molecules crossing clinical phases by generative AI
◦ Scaffolds preserve essential drug-like properties by maintaining the pharmacophores of bioactive compounds
-
Guaranteed Resupply◦ Through robust chemistry development, optimize reaction conditions to consistently achieve reliable results
◦ Comprehensive and thoroughly documented experimental reports ensure high reproducibility
-
In-house Bioassays◦ 50+ target-based assays & 100+ tumor cell lines readily available for in-house screening
◦ Tailored bioassay development to meet specific project needs
How does it work?
XtalPi's Targeted Library Design
Main scaffold and building blocks by XtalPi using Generative AI
◦ Scaffold hopping using AI generative model and expert modification for novel design
◦ Multi-reactive sites enable fast and broad library expansion
Synthesizability Assessment
◦ Physics-based models assess reactivity based on the electronic/steric properties of the reaction center and detect competing sites and side reactions using empirical structural rules.
◦ AI-based models predict synthetic feasibility using machine learning models trained on high-quality experimental data.
XtalPi enumerates libraries with the generated novel scaffolds
◦ Drug-like small molecules through virtual screenings
◦ Enhanced physiochemical properties and ADMET profile
◦ Diversity selection in terms of skeleton, functional groups, property, pharmacophore