Using statistical techniques and replication samples for imputation of metabolite missing values

TL;DR

This paper introduces a statistical approach using replication samples to understand and improve the imputation of missing metabolite data, demonstrating that missing values are not always low and validating KNN as an effective method.

Contribution

The study presents a novel statistical technique leveraging replication samples to characterize missing metabolite data and identifies KNN as an optimal imputation method.

Findings

Missing values are approximately uniformly distributed across metabolites.

Imputation should not assume missing values are always low.

KNN is validated as an effective imputation approach.

Abstract

Background: Data preparation, such as missing values imputation and transformation, is the first step in any data analysis and requires crucial attention. Particularly, analysis of metabolites demands more preparation since those small compounds have recently been measurable in large scales with mass spectrometry techniques. We introduce novel statistical techniques for metabolite missing values imputation by utilizing replication samples. Results: To understand the nature of the missing values using replication samples, we obtained the empirical distribution of missing values and observed that the rate of missing values is approximately distributed as uniform across the metabolite range. Therefore, the missing values cannot be imputed with the lowest values. Using the identified distribution, we illustrated a simulation study to find an optimal imputation approach for metabolites. Conclusions: We demonstrated that the missing values in metabolomic data sets might not be necessarily low value. After identification of the nature of missing values, we validated K nearest neighborhood as an optimal approach for imputation.