Work

Best Practices and Considerations for Applying Multiple Linear Regression in Organic Chemistry Research

Year: 2026

Type: article

Abstract: This Synopsis provides guidance for designing and executing Multiple Linear Regression (MLR) campaigns in organic chemistry. It outlines key steps of a robust workflow for accelerating reaction outcome prediction and maintaining interpretability, including data preparation, feature generation, data distribution analysis, model building, validation, and virtual screening. Emphasis is placed on defining a clear chemical objective and establishing mechanistic hypotheses. Representative examples illustrate how data size and distribution guide data splitting strategies and ratios, ultimately shaping model reliability and interpretability.

Source: The Journal of Organic Chemistry

DOI: 10.1021/acs.joc.5c03206

Authors: Austin LeSueur, Pauline Bianchi, Simone Gallarati, Sarah J. Lefave, Matthew S. Sigman

Institutions: University of Utah, University of California, Los Angeles

Language: English

Topic: Machine Learning in Materials Science

Subfield: Materials Chemistry

Field: Materials Science

Domain: Physical Sciences

Open Access status: green