Since the Bullvine started in February (and even before), I have had many dedicated breeders ask me ‘Why do the Artificial Insemination companies keep going back to the same herds for young bulls when few if any bulls from those herds ever make it to active proven status?”’ That prompted me to ask – is that true?
Answering the tough questions
For many breeders it can be hard to determine which herds they should use young sires from and which ones they shouldn’t. For many AI companies, and sire analysts they have had their thoughts about which herds are hot and which ones are not, but never had the numbers to prove it. This has us asking has genomics truly eliminated the hothouse effect in sire sampling (read more – Has Genomics Knocked Out Hot House Herds). In true Bullvine style we decided to tackle this tough question to help breeders have confidence in which genomic tested young sires to use. We first took a look at all the herds that had more than one sire receive their first official proof in the August 2012 proof run.
Here is what we found:
Herd Count Official GLPI DGV LPI PA LPI GPA LPI
Alta 4 1022 1085 969 1043
Claynook 3 969 982 1018 995
Comestar 3 1436 1368 1507 1418
Gen-I-Beq 5 1658 1701 1800 1737
Gillette 3 771 692 1045 819
Lorka 3 1239 1190 1181 1187
Stanton 6 781 737 1383 970
Velthuis 3 1369 1280 1756 1452
In comparing the columns here is what we found:
Herd Count DGV vs. Official PA vs. Official GPA Vs. Official
Alta 4 63 -53 21
Claynook 3 13 49 26
Comestar 3 -68 71 -18
Gen-I-Beq 5 43 142 79
Gillette 3 -79 273 48
Lorka 3 -49 -58 -52
Stanton 6 -43 603 189
Velthuis 3 -88 387 83
In wanting to see if the Aug ’12 results were just a point in time for these herds or if it was truly reflective, we decided to look at all the bulls proven since the introduction of genomics (Aug 2009) from these same eight herds. Here is what we found:
Herd Count Official GLPI DGV LPI PA LPI GPA LPI
Alta 11 929 936 873 913
Claynook 14 -60 -42 535 166
Comestar 35 749 746 1030 848
Gen-I-Beq 22 1057 1092 1171 1120
Gillette 17 1337 1326 1442 1368
Lorka 7 419 396 749 523
Stanton 27 664 648 1319 889
Velthuis 4 1466 1421 1651 1504
Again comparing the columns here is what we found:
Herd Count DGV vs. Official PA vs. Official GPA Vs. Official
Alta 11 7 -55 -15
Claynook 14 18 595 226
Comestar 35 -3 281 99
Gen-I-Beq 22 35 114 64
Gillette 17 -11 105 30
Lorka 7 -24 330 104
Stanton 27 -16 656 226
Velthuis 4 -45 184 37
Let the numbers do the talking
In studying proven bulls we found that most of them fall within the range for difference between the individual bull’s parent averages and official proof of 150-200 LPI points. And for most of the herds hear falling within that same range.
To go deeper and identify which sires are the best sires to sample, and for breeders to use and which sires should not be sampled or used we took a closer look at how DGV’s, Parent Averages and GPA LPI’s compared to their Estimated Daughter Performance*. Here is what we found:
Herd Count EDP* EDP vs. Official DGV vs. EDP PA vs. EDP GPA vs. EDP
Alta 11 926 -2 9 -53 -13
Claynook 14 -148 -88 106 683 314
Comestar 35 720 -29 26 309 128
Gen-I-Beq 22 1003 -54 89 168 117
Gillette 17 1338 1 -12 104 29
Lorka 7 409 -10 -13 340 113
Stanton 27 607 -56 40 712 282
Velthuis 4 1498 32 -77 153 6
*Please note to calculate Estimated Daughter Performance we took each sire’s official proof and back solved considering his DGV and PA using the published CDN formulas and weightings.
These results are consistent with our previous findings that DGV’s are by far the most accurate indicator of which sires you should use/sample. As well, why you should not use sires that have DGV’s below their parent averages (read more – 7 Reasons Why You Shouldn’t Use Genomic Sires With DGV’s Lower Than Their Parent Averages)
The Bullvine Bottom Line
For both breeders and A.I. companies it can be very challenging trying to figure out which genomic tested sires to use and which breeding programs they should consider investing in. As we have found out the numbers tell the whole truth. Genomic results do give us very reliable information. While it may be true that for some of herds it can be said that they breed better female than male bloodlines. Nevertheless, that does not fully explain why A.I. companies have continued to sample bulls from some herds. It also does not justify why sire sampling herds should be asked to take on the risk in their herd improvement programs. As one of my breeder friends often tells me – use the best, cull or ignore the rest.
Not sure what all this hype about genomics is all about?
Want to learn what it is and what it means to your breeding program?
Thanks for a very interesting article. What I find interesting is the difference between PA and EDP compared to DGV and EDP. If we just had PA for young sires and no genomic data then Stanton looks good 3rd in the list and 300 points behind the top herd, Velthuis. However, with a 700 point drop to the expected daughter performance, they’re now 900 points behind Velthuis. Whereas if we have DGV we see an 800 point difference, which is much closer to the expected 900 point difference.
The use of genomics should mean that the AI companies look at different herds than they have been used to with only Parent Average data. So, the hot house herd effect should be less effective in getting sires sampled.