Similarity measure for sparse time course data based on Gaussian processes

TL;DR

This paper introduces a Gaussian process-based similarity measure for sparse, noisy time course data, improving clustering performance in biological datasets like gene transcriptomics.

Contribution

The paper presents a novel GP-based similarity measure that enhances robustness to noise and is theoretically linked to Euclidean distance under certain conditions.

Findings

GP similarity outperforms traditional measures in noisy, sparse data

Improved clustering results on synthetic and real biological data

Equivalent to Euclidean distance when noise variance is low

Abstract

We propose a similarity measure for sparsely sampled time course data in the form of a log-likelihood ratio of Gaussian processes (GP). The proposed GP similarity is similar to a Bayes factor and provides enhanced robustness to noise in sparse time series, such as those found in various biological settings, e.g., gene transcriptomics. We show that the GP measure is equivalent to the Euclidean distance when the noise variance in the GP is negligible compared to the noise variance of the signal. Our numerical experiments on both synthetic and real data show improved performance of the GP similarity when used in conjunction with two distance-based clustering methods.