Feature impact assessment: a new score to identify relevant metabolomics features in artificial neural networks using validated labels

Introduction Artificial Neural Networks (ANN) are increasingly used in metabolomics but are hard to interpret.Objectives We aimed at developing a feature impact score that is model-agnostic, simple, and interpretable.Methods Feature Impact Assessment (FIA) is calculated by varying combinations of features within their observed value range and checking for changes in prediction outcomes. FIA was implemented in R and tested on metabolomics datasets.Results FIA exceeded LIME and SHAP in selecting biologically meaningful features. Values were comparable across different ANN architectures.Conclusion FIA is a novel score ranking feature impact, helping interpreting ANN in the metabolomics field.

Automated summary

This study aims to develop a model-agnostic, simple, and interpretable feature impact score. Feature Impact Assessment (FIA) is calculated by varying feature combinations and observing changes in prediction outcomes. FIA outperforms LIME and SHAP in selecting biologically meaningful features and is applicable to different ANN architectures.