Predicting spin-crossover behavior in metal-organic frameworks from limited and noisy data using quantile active learning

Spin-crossover (SCO) metal-organic frameworks (MOFs) hold great promise for sensing, spintronics, and gas-related applications, however, only a small number of SCO-active examples are known among the thousands of MOFs already synthesized. Computational screening enhanced by machine learning offers a powerful route to uncover these hidden candidates much more rapidly than trial-and-error experiments. However, progress is limited by the computational complexity of obtaining accurate adiabatic energy differences, as these typically require separate geometry optimizations for both spin states, a process that is technically challenging, prone to convergence failures, and difficult to automate at scale. To mitigate these issues, we introduce a data-efficient strategy based on Quantile Regression Tree-based Active Learning, designed to navigate large chemical spaces while remaining robust to noisy and scarce labels obtained from unrelaxed geometries. After actively selecting a 200-sized subset of representative MOFs for electronic-structure evaluation, a Random Forest regressor trained on this data accurately identifies SCO-relevant candidates despite label noise, recovering ~ 82% of PBE-defined positives based on relaxed-geometry adiabatic energy differences, with only two false negatives. Negative-control tests further indicate that the model captures chemically meaningful trends. Applying the model to the unlabeled dataset yields a new collection of high-confidence SCO MOFs, which we denote pSCO-105. This work shows that SCO-relevant electronic properties can be reliably identified from limited and imperfect data through smart training-set selection, enabling accelerated screening of SCO MOFs.

Authors

Institutions

Publication Details

Journal
npj Computational Materials
Published
2026-11-18
DOI
https://doi.org/10.1038/s41524-026-02365-7
Citations
1
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.84

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Predicting spin-crossover behavior in metal-organic frameworks from limited and noisy data using quantile active learning

Martin Uhrin, Émilie Devijver, Ashna Jose, Roberta Poloni et al.
1 citations
npj Computational Materials
Machine Learning in Materials Science
0.84
article

Predicting spin-crossover behavior in metal-organic frameworks from limited and noisy data using quantile active learning

Martin Uhrin, Émilie Devijver, Ashna Jose, Roberta Poloni, Noel Jakse
article en
1 citations

Abstract

Spin-crossover (SCO) metal-organic frameworks (MOFs) hold great promise for sensing, spintronics, and gas-related applications, however, only a small number of SCO-active examples are known among the thousands of MOFs already synthesized. Computational screening enhanced by machine learning offers a powerful route to uncover these hidden candidates much more rapidly than trial-and-error experiments. However, progress is limited by the computational complexity of obtaining accurate adiabatic energy differences, as these typically require separate geometry optimizations for both spin states, a process that is technically challenging, prone to convergence failures, and difficult to automate at scale. To mitigate these issues, we introduce a data-efficient strategy based on Quantile Regression Tree-based Active Learning, designed to navigate large chemical spaces while remaining robust to noisy and scarce labels obtained from unrelaxed geometries. After actively selecting a 200-sized subset of representative MOFs for electronic-structure evaluation, a Random Forest regressor trained on this data accurately identifies SCO-relevant candidates despite label noise, recovering ~ 82% of PBE-defined positives based on relaxed-geometry adiabatic energy differences, with only two false negatives. Negative-control tests further indicate that the model captures chemically meaningful trends. Applying the model to the unlabeled dataset yields a new collection of high-confidence SCO MOFs, which we denote pSCO-105. This work shows that SCO-relevant electronic properties can be reliably identified from limited and imperfect data through smart training-set selection, enabling accelerated screening of SCO MOFs.

npj Computational Materials
Institut polytechnique de Grenoble (FR), Centre National de la Recherche Scientifique (FR), Science et Ingénierie des Matériaux et Procédés (FR), Laboratoire d'Informatique de Grenoble (FR), Imperial College London (GB), Université Grenoble Alpes (FR)
Agence Nationale de la Recherche, Centre National de la Recherche Scientifique
Openalex Percentile: Top 37%
Machine Learning in Materials Science
0.84
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon