Divergence in the circulating small non-coding RNA transcriptome based on female pig fertility status

The reproductive performance of sows is critical to pork industry productivity. However, genetic improvement of reproductive and litter traits remains challenging due to low heritability and unfavourable genetic correlations. Integrating small non-coding (sncRNA) biomarkers to supplement existing selection protocols may enhance female pig reproductive potential selection. This study aimed to compare circulating serum sncRNA profiles between infertile (INF; nonpregnant after two consecutive artificial inseminations, n = 30) and high reproductive potential (HRP; average lifetime live litter size ≥ 13 piglets, n = 50) female pigs and identify candidate miRNA biomarkers associated with reproductive potential. Small RNA sequencing revealed 19 significantly (P < 0.05) different expressed sncRNAs (12 upregulated and 7 downregulated in INF) and 13 significantly (P < 0.05) different expressed miRNAs (8 upregulated and 5 downregulated in INF). Many of the identified DEmiRNAs have known roles in female reproduction and infertility, however, whether these differences are causal or compensatory responses of female pig reproductive potential level remains unclear. Using a complementary feature selection approach, 22 candidate miRNA biomarkers were selected, including miR-140-3p, miR-215-5p, and miR-7142-3p. The trained support vector machine (SVM) classified female pig reproductive potential with high accuracy (91 %), sensitivity for HRP (96 %) and specificity for INF (83 %), supporting its potential to improve selection efficiency of reproductive potential in the breeding herd. Validation of these 22 candidate miRNA biomarkers and model in a larger, independent cohort of female pigs is necessary to determine generalizability and use for practical application in on-farm breeding herd selection.