Link to the article here.
Upon being approached by Genomic Prediction to write this story, the writer Antonio Regalado agreed with Genomic Prediction to permit us to respond to any specific criticism in the article, as part of a stated commitment to veracity in reporting.
However, this agreement was not upheld, and the writer did not include any of Genomic Prediction's responses or corrections given shortly before the article was published, and furthermore refused to update the article. As a result, the article contains no responses from us, and is highly inaccurate.
Our response was to point out that we do have peer-reviewed publications substantiating our science (none of it referenced in the article) and extensive support. In our correspondence with Antonio Regalado, this substantiation was extensively supplied and cited. Regalado could have chosen to include any one of these factual corrections and responses, which would serve to at least counterbalance the one-sided critique directed against Genomic Prediction in the article, in particular the critique that "the science isn't there".
However, the writer refused to cite any of the facts in our response, and broadly speaking did not balance his one-sided report with a single fact supporting our science. Among the facts given to Antonio were the following:
- We have published peer-reviewed papers evidencing validation of Genomic Prediction's ability to reduce disease risk by selecting between siblings. None of our critics are even published in the field; they are simply speculating, without reference to our published evidence.
- Our testing has been validated in peer-reviewed research to be able to reduce disease risk by selecting between siblings, in validation cohorts using thousands of siblings.
- Our PGT-P predictors are validated to have AUC exceeding 0.9; a fact which makes it impossible to credibly call our predictions 'soft', given that they are significantly harder than many common PGT-M tests (such as BRCA1 and BRCA2).