checkSI: Computer-Assisted Pre-Submission Checking of Characterization Data for Organic Compounds
Year: 2026
Type: preprint
Abstract: Characterization records in organic chemistry link a compound's name with its NMR, high-resolution mass spectrometry (HRMS) and, where available, elemental analysis data. These values are often transferred manually into the Supporting Information (SI). An analysis of Organic Letters SI files found errors of concern in about one in ten compounds with accurate mass data. Here we present checkSI, an open-source tool that checks reporting conventions and consistency among characterization fields. It reads SI files and manuscripts in text-based PDF or Word (DOCX) format. checkSI extracts characterization records in supported formats and checks their completeness. With name-to-structure conversion enabled, it compares the derived molecular formula with HRMS or elemental analysis formulas and NMR signal counts with those expected from the derived structure. Optional DP5q scoring assesses the compatibility of the proposed structure with the reported 13 C NMR data. In 83 published documents, manual review confirmed at least one data inconsistency in 188 of the 2572 recognized compound descriptions (7.3%). Because checkSI runs locally, authors can check unpublished data and correct errors before submission.
Source: ChemRxiv
Institution: Technion – Israel Institute of Technology
Language: English
Subfield: History and Philosophy of Science
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7225148187