R语言中为什么后者比前者快这么多??

 

前者:
for
(row in 1:nrow(gterms)) { gene_terms <- str_split(gterms[row,"GOs"], ",", simplify = FALSE)[[1]] gene_id <- gterms[row, "query"][[1]] tmp <- data_frame(gene = rep(gene_id, length(gene_terms)), term = gene_terms) dim(tmp) gene2go <- rbind(gene2go, tmp)} gene_ids <- egg$query gene_ids eggnog_lines_with_go <- egg$GOs != "" eggnog_annoations_go <- str_split(egg[eggnog_lines_with_go,]$GOs, ",")
后者: gene_to_go
<- data.frame(gene = rep(gene_ids[eggnog_lines_with_go], times = sapply(eggnog_annoations_go, length)), term = unlist(eggnog_annoations_go)) head(gene_to_go)

 

posted @ 2024-03-25 23:10  小鲨鱼2018  阅读(5)  评论(0编辑  收藏  举报