Which Breeding Index Is Right for You?

Thursday, January 26th, 2017

Have you ever let your search for which semen to buy, go beyond the individual sire level? I have been as keen as the next breeder to find individual sires that will do the best job of genetically improving cows. Proof day meant that I would read through a multitude of total merit index listings to find the sires for me. However, this year I decided to put more focus on the end results I want to achieve and use only one index when selecting my initial short list of sires. What follows are my thought processes and calculations.

Indexes – Yes or No?

For me the answer is automatic. I rely on genetic total merit indexes when choosing sires and females also. My reasoning comes down to one factor. By using a total merit index to base my sire selection decisions, I can get 6 to 8 out of 10 sire daughters that meet or exceed my expectations. That compares to 2 to 3 out of 10 when I use raw sire daughter averages or a single trait index (i.e. milk yield or udder composite). This means triple the number of cows that stay around for a third lactation and those are the odds worth working with. Yes, we all have read on the Milk House that some breeders are willing to throw away nine of ten daughters of a sire to get that very outstanding one for production or type. But for most of us, that is too costly an alternative.

Total Merit Indexes – Many are Available

Every dairy breeding country has its own total merit index. I have just researched and found twenty such indexes, and there is considerable range amongst them in the emphasis placed on a host of individual traits.

Globally there is often reference to five US combined trait indexes that range from indexes for only type all the way to an index focusing on cheese yield. For this article, the Bullvine has analyzed five American combined or total merit indexes for the current top twenty active proven Holstein sires. Only sires with at least 90% REL were included so that comparisons are based on proven in herd performance.

Let’s Compare the Results of Five Indexes

Averaging the results from the top twenty proven sires shows the results breeders can expect from using a single combined trait index but allows breeders to select or eliminate individual sires per their needs.

Knowing that breeders can range in their breeding goals from highest type to highest production to highest for functional traits, the Bullvine selected PTAT, Fat + Protein Yield, TPI, NM$ and CM$ as the indexes to study.

The assumption is that if breeders use an average sire of one of these five combined indexes they will get the results shown in Table 1. Of course, no breeder uses an average sire but, if one of these indexes is used over a period, the result will closely align with the table values.

Table 1 – Average Proof for Top Twenty US Proven Holstein Sires (Dec ’16)

Trait / Index	Highest Type	Highest Production	TPI	NM$	CM$
	(PTAT)	(Fat + Protein Yield)
Fat + Protein	38	135	129	127	123
Milk	479	1709	1669	1422	1418
DPR	-0.8	0.9	1.4	1.9	1.8
LIV	-2.5	0.5	0.6	2.3	2.1
SCS	2.91	2.94	2.91	2.89	2.89
PL	0.9	4.2	5.2	6.5	6.6
DCE	6.9	5.1	4.9	4.9	4.8
UDC	2.62	0.99	1.38	1.09	1.14
FLC	2.21	1.09	1.37	0.87	0.95
Stature	3.49	0.79	1.89	-0.01	0.28
PTAT	3.15	1.39	1.53	1.19	1.25
TPI	2108	2497	2568	2533	2539
NM$	229	713	725	760	756
CM$	220	729	721	776	783

Color Codes

Highest / Best – Blue

Lowest / Worst – Green

The Bullvine’s synopsis of the results is:

Highest type (PTAT) sires produce below breed average results for yields, fertility, health, longevity and daughter calving ease. For production and functional traits, the results are significantly behind other indexes. As expected the average PTAT is very high at 3.15. Note that very high type is closely associated with extremely tall Holsteins (Stature 3.49). Also, noteworthy is the -2.5 for cow Breeding for type does not equate to easier calving cows or improved cow longevity.
Highest production (F+P) at 135 lbs is very high with breed average values for fertility and livability and okay values for overall type (note stature is about breed average). Additionally, the values for TPI, NM$, and CM$ for this group are much higher than for the highest type sires.
As expected most top TPI sires can be expected to give high averages for most values. Type at almost 1.5 standard deviations above the rolling breed average would result in significant continued improvement for type including very tall Production is also high. However, fertility and cow livability are barely above average, so breeders could not expect to significantly improve these areas that are today considered to be in need of improvement in Holsteins.
The highest NM$ sires will produce daughters that are high or excel for all values except for type. PTAT at 1.20 is above average, but the -0.01 for stature will result in decreasing stature as today’s Holstein average proven sire stature index is likely between 0.6 and 0.8.
The highest CM$ sires will produce very similar results to the most top NM$ sires. So, either NM$ or CM$ could be used in selecting sires.
SCS do not differ amongst the five groupings.
UDC and FLC are at least average for all five groupings. It should be remembered that UDC and FLC indexes are not as useful as the individual linear indexes for udders and feet and legs when mating sire and cow.
PTAT and TPI are not the indexes to use for breeders planning to decrease stature and improve health and fertility in their herds.
Breeders interested in following the NM$ or CM$ approach to breeding may wish to read the Bullvine article ”She Ain’t Pretty, She Just Milks That Way.”

Beyond the US Daughter Proven Holstein Sires

The Bullvine also studied the current top 10 Canadian Holstein proven sires and top 10 US Jersey proven sires and found very similar results to the proven US Holstein sires. In Canadian Holsteins, breeders should use Pro$ as the equivalent to NM$ or CM$. In US Jerseys JPI and CM$ gave almost identical results.

For breeders interested in only using top NM$ genomically evaluated Holstein sires, our study shows there are 27 sires above 900 NM$. And inbreeding is not likely to be a problem as they are sired by eleven different sires (4 Altaspring, 4 Montross, 4 Silver, 3 Jedi, 3 Rubicon, 3 Yoder, 2 Supershot, 1 Delta, 1 Josuper, 1 Octoberfest and 1 Pety).

Other Indexes – Now Available

The Bullvine recently produced an article, “Will Genetic Evaluations Go Private?” on other total merit indexes that private breeding companies are publishing. More of these private indexes can be expected in the future. Currently, the Ideal Commercial Cow Index, ICC$ ™, is being used by some breeders and DWP$ ™ sire rankings (Read more: The Complete Guide to Understanding Zoetis’ New Wellness Traits – CLARIFIDE® Plus) have just been published since the latest official proof release in December (See complete lists at Sire Proof Central).

Only You Can Decide

There is no combined trait or total merit index that is perfect for all breeders to use when short listing sires for their herd. Some farms sell their milk to fluid processors; some sell to cheese makers, some processors make powder, and some farms bred for breeding stock sales. There are indexes to address all situations.

If a breeder is hoping to produce bulls for use in A.I. on a significant scale, then using an index that commercial dairy farmers would us is an advantage when breeding to produce breeding bulls. A high genomic value is also critical when selling bulls to A.I.

The Bullvine recommends that all breeders have a plan (Read more: What’s the plan?) based on goals and expected results.

The Bullvine Bottom Line

Knowing the desired end result, five years down the road, from a breeding program is a prerequisite to selecting sires. The majority of tomorrow’s dairy cattle breeders will need to be farm profit focused. My first-choice total merit index, for selecting sires, is now NM$.