Expert in the Loop AI – Polymer Discovery

Format	CSV
License	CDLA-Sharing
Domain	Sequence Classification
Number of Records	125,143
Data Split	118 accept and 125143 reject sequences
Size	14 MB
Author	Nathaniel H. Park, Dmitry Zubarev
Data Origin	IBM Research
Dataset Version	1.0.0
Dataset Coverage	N/A
Business Use Case	Materials Discovery: This dataset can be used to assess if a given cyclic lactone is suitable for experimental investigation. Adjudication accounts for the expert-level judgement about the practicality of the lactone synthesis (expectation of the low cost, short synthetic route, and high yield) and utilization as a monomer (no side-reactions, reasonable reactivity under polymerization conditions).

Sequence	Adjudication
NC(N)=NC(=O)CC1COC(=O)OC1	accept
CC(CNC(=NC(=O)On1cc[n+](CCCCCCCOC=O)c1)Nn1cc[n+](C)c1)C1COC(=O)OC1	reject
O=C1OCC(CC2CCCOC2=O)CO1	accept
C=C(C)C(=O)CC1COC(=O)OC1	accept

Feature	Description
sequence	A candidate cyclic lactone.
adjudication	The expert-level judgement about the practicality of the lactone synthesis. Accept indicates the candidate is suitable for further experimental investigation.