After estimating the number of genes involved in TD risk, we utilized this number to predict the gene discovery yield, as additional TD trios are whole-exome sequenced. We fixed the gene number at 420, and varied the cohort size. Therefore, we calculated the number of variants in each iteration based on the observed mutation rate in probands (based on 199 de novo damaging mutations – 7 that failed confirmation = 192 variants). As was done in the MLE, we randomly selected TD risk genes, and then assigned a fraction of these variants to TD risk genes and the remaining fraction to non-TD risk genes (see below for estimation of fractions). We performed 10,000 permutations at each cohort size, and randomly generated LGD and Mis3 variants separately, using their respective rates and per gene likelihoods. We then combined these data and each permutation was run through the TADA de novo algorithm with the same parameters used above for the observed data to assess the per gene q-values. We then recorded the number of probable genes (q < 0.3) and high confidence
