Genes and Mechanisms Associated With Experimentally Induced Bovine Respiratory Disease Identified With Supervised Machine Learning Methodology on Integrated Transcriptomic Datasets

Abstract

Bovine 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.

Hematological and gene co-expression network analyses of high-risk beef cattle defines immunological mechanisms and biological complexes involved in bovine respiratory disease and weight gain

“…Metaphylactic use of antimicrobials at arrival reduces risk of morbidity and mortality across beef production systems, however this management practice is variable in efficacy, in both rates of disease across cattle populations and in pharmacological choice, and the practice is suspected to drive expansion of antimicrobial resistance, a growing societal concern [52,57,58]. Given this background, our research group and others have focused on evaluating host transcriptomes at arrival, to better characterize host-driven mechanisms and develop candidate mRNA biomarkers associated with clinical BRD outcomes [18,19,20]. These studies have provided valuable information regarding cattle treated based on clinical signs of BRD, but these studies heavily rely on semi-objective evaluation of BRD cases and may miss underlying subclinical or misdiagnosed disease.…”

“…Based on our previous work, it can be inferred that host gene expression captured at facility arrival is variable across BRD severity cohorts [20,40,41]. Therefore, we assessed whether the at-arrival co-expression patterns and modules found in this study were well preserved across an RNA-Seq data set from an independent population of cattle.…”

“…The gene-level raw count matrix generated from our previous research was utilized for this study [20]. Briefly, RNA was isolated via Tempus Spin RNA Isolation Kits (Thermo Fisher Scientific; Waltham, MA, USA), following manufacturer's protocol.…”

“…When evaluating the relationships between cytokine gene expression and CBC data in cattle with concurrent BRD, Lindholm-Perry and colleagues discovered that cattle with BRD possessed a comparative increase in numbers of neutrophils, decrease in numbers of basophils, and increased expression of CCL16, CXCR1, and CCR1 [17]. Recent RNA sequencing studies, performed by both our group and others, have identified mechanisms and candidate biomarkers in whole blood associated with BRD development [18,19,20]. However, these studies primarily sought to identify differentially expressed genes (DEGs) between cattle that were or were not treated for BRD based on clinical signs.…”

“…This analysis approach falls into the field of systems biology, where, in contrast to reductionist biology, molecular components are pieced and scaled together to better understand disease and generate novel hypotheses [23,24]. In this respect, we sought to build networks of co-expressed genes, utilizing the full structure of previously published gene expression data [20], and discover relationships between gene expression and cellular hematological components, which may elucidate and/or further confirm genes and mechanisms related to BRD development or resistance.…”