ETS translocation variant 1 (Ets-related protein 81) [ER81]

PublicationsПравить

Survival analyses in Holstein cows considering direct disease diagnoses and specific SNP marker effects.

The aim of the present study was to infer influences of health disorders and of specific SNP markers on longevity in Holstein cows via survival analyses. Longevity was defined as the length of productive life (LPL), reflecting the interval from first calving until the culling or censoring date. In this regard, we considered longevity records from 129,386 Holstein cows from 57 large-scale herds, with calving dates in first parity between January 2004 and December 2017 (30.27% censored records). We selected specific diseases from the overall categories claw disorders, udder diseases, metabolic disorders, and female fertility disorders within 3 stages of parities 1, 2, and 3. Lactation stage 1 was the period from calving to days in milk (DIM) 59, lactation stage 2 from DIM 60 to DIM 299, and lactation stage 3 from DIM 300 to the next calving date. The effects of the diseases on culling risk ratios were estimated via Weibull proportional hazards models. In this regard, we used 3 different modeling strategies. In modeling strategy M1S, binary diseases from different parities and lactation stages for the same diagnosis were modeled as explanatory variables in separate runs. Modeling strategy M3S included diseases for the same diagnosis from stage 1, 2, and 3 in the same parity simultaneously. Modeling strategy M9S implied consideration of diseases for the same diagnosis from different lactation stages and from all 3 parities simultaneously. The effect of the same diseases on culling risks from M1S and M3S were similar, with increasing detrimental effect of diseases recorded in later lactation stages and parities. The strongest disease effect on LPL was detected for clinical and subclinical mastitis recorded in the middle of the third lactation, with culling risks of 2.59 and 2.40, respectively, and for claw disorders from the last stage in third lactation (culling risks in the range from 1.85 to 2.29). The effective (ignoring the proportion of censoring; h ) and equivalent (considering proportion of censoring; h ) heritabilities for LPL when considering diseases from specific stages in parities 1 and 3 were quite low ( h 0.02 - 0.17; h 0.01 - 0.12). Simultaneous consideration of same disease diagnoses across stages (M3S) and across lactations (M9S) contributed to a LPL heritability increase. Ignoring diseases as explanatory variables in survival analyses was generally associated with a decline of genetic LPL variances and LPL heritabilities. A genome-wide association study for LPL was based on estimated de-regressed proofs from 17,362 genotyped cows. Six SNP located on Bos taurus autosomes 1, 4, 10, 13, and 28 were significantly associated with LPL (using a 5% false discovery rate). Gene annotations via Ensembl identified the 4 potential candidate genes ETV1, ONECUT1, MACROD2, and SIRT1, which directly (via disease resistance mechanisms) or indirectly (via milk productivity) influence dairy cow longevity. Genotypes of the 6 significantly associated SNP were considered as fixed effects in Weibull hazards models, but their effects on culling risks were nonsignificant. Heritabilities for LPL from all SNP-based survival models considering the single SNP separately or the 6 SNP simultaneously were 0.05 (h ) and 0.12 (h ). The small number of significantly associated SNP in genome-wide associations and the minor effect of specific SNP on LPL in survival analyses underline the polygenetic nature of longevity.


Keywords

  • SNP effect
  • Weibull hazards model
  • genetic parameter
  • health disorder
  • longevity