Fig. 1: Flow chart of the analysis conducted in the discovery population (Bos indicus cattle).

Analysis was performed in order a to g as follows: a Single trait genome-wide association studies for four fertility traits (DTC, AFC, preg_st, wks_preg) were performed using 31 million imputed whole genome sequence (WGS) variants. b A stepwise conditional multi-trait GWAS (CMT-GWAS) analysis was performed for fertility traits (DTC, AFC, preg_st, wks_preg). c Cis-eQTL discovery was performed for a sample of 489 heifers and cows with RNA-seq and imputed WGS data. d The most significant cis-eQTLs for each eGene (n = 4376, FDR < 0.01) were selected. Additionally, we added 225 genome-wide significant GWAS variants from the single trait and CMT-GWAS model described above, applying a significance threshold P ≤ 5 × 10−8. We also added variants within 100 bp either side of each selected variant. e Trans-eQTL mapping for significant variants identified through cis-eQTL or GWAS analysis. We specifically targeted genes where the variant was located more than 5 Mb away from the gene on the same chromosome. f Summary-data-based mendelian randomisation (SMR) analysis was performed by integrating GWAS with cis-eQTL results. g Follow-up study of the final combined list of significant genes associated with fertility in cattle.