DTW-MIC Coexpression Networks from Time-Course Data

TL;DR

This paper introduces DTW-MIC, a new similarity measure for coexpression networks from time-course data that captures non-linear interactions and time shifts, improving network modeling accuracy.

Contribution

The paper proposes DTW-MIC, combining MIC and DTW, to enhance coexpression network inference from time-series data, addressing limitations of PCC.

Findings

DTW-MIC outperforms PCC in synthetic datasets.

DTW-MIC provides more accurate network reconstructions.

Effective on transcriptomic data.

Abstract

When modeling coexpression networks from high-throughput time course data, Pearson Correlation Coefficient (PCC) is one of the most effective and popular similarity functions. However, its reliability is limited since it cannot capture non-linear interactions and time shifts. Here we propose to overcome these two issues by employing a novel similarity function, Dynamic Time Warping Maximal Information Coefficient (DTW-MIC), combining a measure taking care of functional interactions of signals (MIC) and a measure identifying horizontal displacements (DTW). By using the Hamming-Ipsen-Mikhailov (HIM) metric to quantify network differences, the effectiveness of the DTW-MIC approach is demonstrated on both synthetic and transcriptomic datasets.