In the United Kingdom, says Sue Kenwrick, a principal genetic counselor at Addenbrooke's Hospital in Cambridge and chair of the training panel of the Association of Genetic Nurses and Counsellors (AGNC), "my experience … is that referrals are going up rapidly and genetic counseling roles are growing and growing, and there are not enough of us … to meet workforce needs".
Preventive newborn care is covered in iCCM and basic VHT training (Panel 1).
As expected, we observed that accuracy of GEBV increased with increased relatedness between the animal and the training panel.
Approximately 10% of the animals were randomly assigned to a validation panel (V) in which predictions would be made, whereas the remaining 90% were used as the training panel (T) to estimate the parameters necessary for prediction.
To further investigate the effect of relatedness between training and validation animals, we selected the youngest 87 animals (approximately 10% of the population) that included 82 animals with at least a sire or a grand-sire in the training panel.
We assembled a training panel of breast cancer tumour blocks from 67 BRCA1, 71 BRCassociatedted and 105 sporadic young onset cases (≤ 40 years at diagnosis) from the Prospective study of Outcomes in Sporadic versus Hereditary breast cancer (POSH) study.
We investigated this assertion in two ways: (1) by computing average r GEBV for animals with different number of relatives in training panel and (2) by regressing r GEBV on the average top 10 relatedness in the genomic relationship matrix.
Our results show that while it is true that individuals with close relatives in the training panel will have on average more accurate GEBV, the individual accuracies obtained from the G matrix would be overestimated.
We implemented two separate imputation experiments that differ in the size of the high-density reference panel used for imputation: (1) a reference panel of 128 Yorkshire haplotypes or (2) a reference panel combining the 128 Yorkshire haplotypes with the haplotypes of all animals that are part of the training panel (~1700 additional haplotypes) in the respective cross-validation dataset.
We found that increasing the number of close relatives from one to four in the training panel increased average r GEBV by about 0.1 decimal points across the three traits in this study (from an average of = 0.63 to = 0.73 regardless of the trait considered).
