Co-efficient of Inbreeding Study

Remember that a COI of 3.125% represents the equivalent of a single 3x3 or two 4x4 or four 5x5 inbreedings. Likewise a COI of 0.78 represents a single 4x4 inbreeding and a COI of 0.19 a single 5x5 inbreeding.

I Calculated the coefficient of inbreeding to 6 generations for every horse in my database from the mid 90's until 2002 that I had a complete 6 generation pedigree for. These were split into 2 categories: Group winners and Others.

COI	Group	Other	Total
< 1%	5207	96639	101846
< 2%	2098	45257	47355
< 3%	699	15421	16120
< 4%	272	6240	6512
< 5%	99	2688	2787
< 6%	26	551	577
< 7%	18	700	718
< 8%	5	435	440
< 9%	1	96	97
Total	8425	168027	176452

Null hypothesis is that regardless of COI there is no difference in results
Using Chi²the result is 83.0
Degree of Freedom 8
Chi²variate for 8df at 99.5% is 21.96
Conclusion is that there is a significant difference in results. Also the category that provides by far the most likely successful outcome are those with a COI less than 1%.

Narrowing the same categories down further:

COI Group Other Total

< 0.5 2856 48656 51512

< 1.0 2351 47983 50334

< 1.5 1380 29643 31023

< 2.0 718 15614 16332

< 2.5 492 10777 11269

< 3.0 207 4644 4851

< 3.5 139 3023 3162

< 4.0 133 3217 3350

< 4.5 74 1938 2012

< 5.0 25 750 775

< 5.5 16 361 377

< 6.0 10 190 200

< 6.5 5 229 234

< 7.0 13 471 484

< 7.5 5 300 305

< 8.0 0 135 135

< 8.5 1 76 77

< 9.0 1 25 26

Total 8426 176458 184884

Using Chi²the result is 126.70
Degree of Freedom 17
Chi²variate for 17df at 99.5% is 35.72
Same conclusion as above.
Also the < 0.5% category is far and away the most successful category.

The largest category is by far the < 1% which contains more results than all other categories combined. Breaking down just that category:

COI Group Others Total

< 0.1 513 7833 8346

< 0.2 540 8306 8846

< 0.3 638 10314 10952

< 0.4 546 10929 11475

< 0.5 619 11274 11893

< 0.6 572 11230 11802

< 0.7 547 10374 10921

< 0.8 428 9593 10021

< 0.9 400 8772 9172

< 1.0 404 8014 8418

5207 96639 101846

Using Chi²the result is 80.0
Degree of Freedom 9
Chi²variate for 9df at 99.5% is 23.59
Conclusion is even in the less than 1% category the less inbreeding the better.

but let us not get carried away .. knowing there is a significance is not quite the same as knowing why and how we can use it to our advantage.

One thing I do know is that something very important happens when there are very low levels of inbreeding between sire and dam. Hybrid vigour strikes again!