The use of the probit model to predict pregnancy status of buffalo based on physio-chemical properties of estrual mucus

Improper timing of artificial insemination with respect to ovulation is one of the major factors hampering the conception rate in buffalo. The present study was an attempt to relate physio-chemical changes in estrual mucus to subsequent pregnancy status in order to find their optimal values for determining the time for artificial insemination (AI). Serum estradiol, total protein and dry matter contents of estrual mucus were evaluated to predict the subsequent pregnancy in 36 buffalo during October 1988 to February 1989. Serum estradiol was determined by radioimmunoassay (RIA); spinnbarkeit, dry matter and total protein were determined by standard methods. Multivariate probit analyses were carried out to relate these variables to subsequent pregnancy status. Elasticity and protein concentration were significantly related to prediction probability of pregnancy status, and they predicted the pregnancy status 86% of the times correctly (P < 0.05). The probability of pregnant animals being correctly classified was 0.76, whereas the corresponding value for non-pregnant animals was 0.95. The present study demonstrated the possibility of using such a statistical model on mucus characteristics for determining proper AI time for better conception rates in Nili-Ravi water buffalo.