Genes and Mechanisms Associated With Experimentally Induced Bovine Respiratory Disease Identified With Supervised Machine Learning Methodology on Integrated Transcriptomic Datasets
Institutional Repository Document
Overview
Research
Identity
Other
View All
Overview
abstract
AbstractBovine respiratory disease (BRD) is a multifactorial disease involving complex host immune interactions shaped by pathogenic agents and environmental factors. Advancements in RNA sequencing and associated analytical methods are improving our understanding of host response related to BRD pathophysiology. Supervised machine learning (ML) approaches present one such method for analyzing new and previously published transcriptome data to identify novel genes and mechanisms. Our objective was to apply ML models to lung and immunological tissue datasets acquired from previous clinical BRD experiments to identify genes that classify disease with high accuracy. Raw mRNA sequencing reads from 151 bovine datasets (n=123 BRD, n=28 control) were downloaded from NCBI-GEO. Quality filtered reads were assembled in a HISAT2/Stringtie2 pipeline. Raw gene counts for ML analysis were normalized, transformed, and analyzed with MLSeq, utilizing six ML models. Cross-validation parameters (5-fold, repeated 10 times) were applied in a 70:30 training/testing ratio. Downstream analysis of genes identified by the top sparse classifiers for each etiological association was performed within WebGestalt and Reactome (FDR < 0.05). Support vector machines was routinely the top non-sparse classifier for predicting etiological disease versus sham control. Nearest shrunken centroid and Poisson linear discriminant analysis with power transformation could reliably classify IBR and BRSV with 100% accuracy. Genes identified in IBR and BRSV, but not BVDV, were related to type I interferon production and IL-8 secretion, specifically in lymphoid tissue and not lung. Genes identified in Mannheimia haemolytica infections were involved in activating classical and alternative pathways of complement. Novel findings, including expression of genes related to reduced mitochondrial oxygenation and ATP synthesis in consolidated lung tissue, were discovered. Genes identified in each analysis represent distinct genomic events relevant to understanding and predicting clinical BRD. The few genes shared across analyses may be reliably associated with clinical BRD. Our analysis demonstrates the utility of ML with published datasets for discovering functional information to support prediction and understanding BRD acquisition.
