Fish ecotyping based on machine learning and inferred network analysis of chemical and physical properties

Feifei Wei,Kengo Ito,Jun Kikuchi,Kenji Sakata,Yasuhiro Date,Taiga Asakura

doi:10.1038/s41598-021-83194-0

Abstract

Functional diversity rather than species richness is critical for the understanding of ecological patterns and processes. This study aimed to develop novel integrated analytical strategies for the functional characterization of fish diversity based on the quantification, prediction and integration of the chemical and physical features in fish muscles. Machine learning models with an improved random forest algorithm applied on 1867 muscle nuclear magnetic resonance spectra belonging to 249 fish species successfully predicted the mobility patterns of fishes into four categories (migratory, territorial, rockfish, and demersal) with accuracies of 90.3–95.4%. Markov blanket-based feature selection method with an ecological–chemical–physical integrated network based on the Bayesian network inference algorithm highlighted the importance of nitrogen metabolism, which is critical for environmental adaptability of fishes in nutrient-rich environments, in the functional characterization of fish biodiversity. Our study provides valuable information and analytical strategies for fish home-range assessment on the basis of the chemical and physical characterization of fish muscle, which can serve as an ecological indicator for fish ecotyping and human impact monitoring.

Highlights

Functional diversity rather than species richness is critical for the understanding of ecological patterns and processes
Machine learning models run with the improved random forest (RF) algorithm based on 1867 muscle nuclear magnetic resonance (NMR) spectra for 249 fish species successfully predicted the mobility patterns of fishes into four categories with accuracies of 90.3–95.4%
Fish muscle was the focus of this study because it is the main edible part of fish and reflects the athletic ability of fish, which is closely related to the ecological characteristics of their habitat and life habits[14]; muscles are the main protein source of fish

Summary

Introduction

Functional diversity rather than species richness is critical for the understanding of ecological patterns and processes. This study aimed to develop novel integrated analytical strategies for the functional characterization of fish diversity based on the quantification, prediction and integration of the chemical and physical features in fish muscles. This study aimed to develop novel integrated analytical strategies for the functional characterization of fish diversity based on the quantification, prediction and integration of the chemical and physical variations in fish muscles (Fig. 1). Ecological category-dependent metabolic networks of the machine-learned chemical features and Markov blanket-based feature selection for an ecological–chemical–physical integrated network were established with the BN inference algorithm to mine the functional connections among the high-dimensional data factors. Our study provides valuable information and analytical strategies for fish home-range assessment on the basis of the chemical and physical characterization of fish muscle, which can serve as an ecological indicator for fish ecotyping and human impact monitoring

Objectives

Methods

Results