Each d ∈ D has a topic proportion as computed in line (1) of Algorithm 2. In Algorithm 3, we simply normalize the topic proportions w i so that for a given D, ∑ i = 1 n w i = 1.
These proportions were computed in a way similar to the calculation of exon frequency (i.e., the proportion of transcript isoforms that contain both [copresent] or neither [coabsent] of the two neighboring ASEs).
The proportion of ROH in each size class was computed in all 104 sequenced individuals.
In each dataset, we computed the proportion of improvement in terms of AUC brought by iASeq over the best single-dataset based ranking method (i.e., dAUC = AU C iAseq − AU C bestsingle AU C bestsingle ).
Then we computed the proportion of tumours in each subgroup in each of the 7 clinical categories.
In addition, we computed the proportion of socioeconomic characteristics for interviewees who participated in both the screening camp and the household survey.
To examine the importance of tandem duplication for generating new miRNA genes in the genome, we also computed the proportion of genes clustered in each genomic region.
For each SNP, we computed the proportion of analyses in which the SNP was selected for inclusion in a final QTL model (RMIP; results for days to anthesis are presented in Figure 5, results for other traits are presented in Supplementary Figure S6).
For smokers, we computed the proportion of smoking in the various settings according to type of enclosure.
Then we counted the number of COGs lost in common and computed the proportion that it represents from the set of COGs lost by at least one bacterium.
