HIGHLIGHTS
SUMMARY
The authors employed the cohort of 841 patients to identify variants most associated to COVID-19 severity which the authors used, along with clinical co-variates such as age and sex, to train and test supervised binary classifiers of severity. Each fold was constituted by a training set, corresponding to 80% of the dataset, which was also employed for variant screening and a remaining 20% for the testing set. For each split, the authors generated a feature matrix for the training set by assigning the allele counts of each screened variant for each sample . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.