Wednesday, March 11, 2015

What should we be learning from recent crossbreeding experiments?

From the May June 2011 Dairy Route Newsletter

Why does anyone crossbreed?       
     (1)   To gain added performance from “hybrid vigor” [heterosis].
What is “hybrid vigor”? 
(2)     The performance response we get from introducing some “warm”[outcross] genes into a “cold”[closed] gene population.
 Why do we wish to change the performance [related behavior] of our animals?     
(3)     Because current generations no longer respond as well as prior generations to trait selection as practiced within our breed.

Let’s stop right there.     “Crossbreeding” is tried when after several generations of gaining performance from standardized methods of selection [which today we call “index”] we no longer see any incremental financial gains—ie, either cow life gets too short, or milk yields no longer grow, or incremental costs are greater than incremental gains.

The problems that grow from believing “change” does not keep changing

The current fixation with “indexing” for commercial herd propogation is based on a “race memory”—the window in time that was the 1960s, when AI studs shifted from bloodline sire selection to progeny data sire selection.     While this was done in a “purebred” context, the net result was we began a mating process in which we “crossed” our herds—we bred “type” cows to “milk” bulls, we bred linebred cows to outcross bulls.   We saw large leaps in productivity from the cow to the heifer generation.   We gave the +1000 PDM figure 99% of the credit for this gain, and assumed this process would work forever.

The 1960s was when all the “single trait selection” ideas gained traction—ideas that later research have uniformly proven to be laden with weaknesses—yet here we are in 2011, and we remain more in love with “single trait selection” than ever, only instead of “PD Milk” we look at “Genomic Net Merit”.

What goes wrong with single trait selection

The problem with single-trait selection [any selection system in which you say, “no matter what, every sire I use must be plus x pounds of milk/fat/protein”] is that you tend to be going back to the same gene wells over multiple generations.    Over time, the underlying gene possession of your cows becomes the same as the gene possession of your mating sires.     For some limited desired genes, a “homozygous” [identical genes in each pairing] pattern is OK—as in gene pairings that stimulate milk yield.   But for too many genes, a “heterozygous” pairing [maintaining the basis for a “hybrid vigor” mating response] is more desirable—these being genes for traits that we need but are not naturally associated with “milk” genes in the mainstream of our breed.   

In other words, think of the typical experiences we have had for the “highest milk” bulls:
(1960) =  If he is +2000m he will be minus for udders and/or feet and legs.
(1970) =  If he is +2000m he will be minus three or four points on butterfat%.
(1980) =  If he is +2000m he will be calving ease and I will end up with small hard calving cows.
(1990) =  If he is +2000m he will be minus for Productive Life [sire short herdlife].
(2000) =  If he is +2000m he will be too high on expected future inbreeding.
(2010) =  If he is +2000m he will be minus for Daughter Pregnancy Rate [sire low natural fertility].

In each generation there would be an exception close to the top, but in our fixation with “number one” it was always possible to overlook him.    Then we would reap the negative effects of trusting only “rank”.

From generation to generation, the negative effects of focusing on one main performance trait tend to accumulate, and this accumulation reflects the trait weakness(es) of the ranking sire line dominant in that generation.     This accumulation is more dangerous if a certain physique is preferred, a further limiting of gene pools by qualitative as well as quantitative selection, leading to extreme frail physiques.

Thus, graziers (who felt they needed shorter, more mobile cows) and large confinement herd managers (who were looking for easier calving heifers, more fertile cows, physiques better adapted to the facility limitations, and more health in general) both pursued crossbreeding as solutions to different problems.

But in fact, if by crossbreeding the bull selected was defined merely by his breed, there is no guarantee of “heterosis” [many Jersey bulls are just smaller, brown Holstein gene accretions] [most Euro Red bulls are proven in single trait ranking “index” systems very similar to our own].   

Whether crossbreeding or outcrossing, the principles remain the same

In any multiple generation analysis, the conclusions remain the same – there is no sustainable shortcut for the full breeding process, IF your goal is to gain on both herd equity and herd net income.

The insurmountable weakness of crossbreeding as most have practiced it is that over time, milk yields will decline.     “Milk” genes get replaced with “Beef” genes; “Cow capacity” genes get replaced with “small” genes; Uniformity for adaptation to group feeding and handling is lost in random variation.    
Basically, because we are not taught that the complexity of gene selection is manageable, and that any short cut method (index selection) (crossbreeding) (linear mating) has declining effectiveness after only three cow generations, we overlook some simple alternatives.

What could have been done (but usually was not):

  We could limit our breed inclusions to one or two breeds, rotating back to our “base” performance breed every second or third generation.    This avoids the loss of productivity, but if we also focus our outcross selection on sires in breed #2 and/or #3 on changing the weaker traits of our base breed, there will be more of a true “heterosis” response than if we pick the crossbreeding sires “the same way” as we pick our performance breed sires.      You have to break the cycle of “likes to likes” matings.

We could stick with our “pure” breed, but open up to using the very sires we usually turned down, the “not enough milk” sires others found desirable—due to their possession of the very traits we had been giving up consistently in our pursuit of “single trait” [milk] selection.    Treat these as an “outcross”, ie, return to your “line” in the following generation.    This is how you find heterosis within any breed.   It is your only way to avoid “inbreeding” today within any breed.

We could go beyond the simply statistical (quantitative) level of bull “proofs” and include a process for managing the physical (qualitative) level of mating— as we do with the “aAa” breeding guide.   If you are unwilling to give up the index ranking selection (“likes to likes”) the qualitative mating process still will impact on exactly those areas of physical adaptation that lead so many to crossbreed initially.

The more successful crossbreeders I know have usually found that at a certain point, saving their own bulls representing the more successful crossed combinations, is more effective than to continue to add new breeds.     [Look at the Kiwicross sires in the latest LIC New Zealand sire directory for examples.]



Crossbreeding assumes we get hybrid vigor because different breeds means different genes.    In fact, under parallel selection systems, all breeds share many gene patterns in common.    The composite (Euro Red) breeds actually offer less heterosis.

Outcrossing within a breed assumes we get hybrid vigor because different bloodlines mean different genes.     In fact, within most breeds, even more gene patterns are in common than is true in crossbreeding, as a result of highly focused trait selection.

“Hybrid vigor” today is thus a result of knowing how to avoid “likes to likes” mating, whether within a breed or crossbreeding.    More heterosis is available when we add in the qualitative level of [physique-related] gene activity in our mating process, and give up arbitrary selection levels on a “primary” trait to acquire higher levels of performance in all our “secondary” traits—leading us to the “different” bulls.

No comments:

Post a Comment