r - 合并行连接由逗号-R 编程分隔的列之一中的内容

Question

我需要帮助合并数据 (mydf) 中具有相同名称的行(即起始列)并连接“ALT”列中的内容，从而根据起始列中的相似值删除所有重复行。我想合并行并连接由逗号分隔的“ALT”列中的内容，并获得如下所示的结果。感谢您的帮助。

> mydf
         chr          start      end REF ALT         TYPE refGene               
       chr10 chr10:176131 176131   C   A          snp nonsynonymous SNV 
       chr10 chr10:159149 159149   C   G snp:17659149 nonsynonymous SNV 
       chr10 chr10:159149 159149   C   T snp:17659149 nonsynonymous SNV 
       chr10 chr10:241469 241469   T   C          snp          splicing 

> result
         chr          start      end REF ALT         TYPE refGene                   
       chr10 chr10:176131 176131   C   A          snp nonsynonymous SNV 
       chr10 chr10:159149 159149   C   G,T snp:17659149 nonsynonymous SNV 
       chr10 chr10:241469 241469   T   C          snp          splicing 

DPUT 在这里:

structure(list(chr = c("chr3", "chr3", "chr3", "chr3"), start = c("chr3:75786036", 
"chr3:75786337", "chr3:75786337", "chr3:75788226"), end = c(75786036, 
75786337, 75786337, 75788226), REF = c("A", "G", "G", "C"), ALT = c("G", 
"A", "T", "A"), TYPE = c("snp:75786036", "snp:75786337", "snp:75786337", 
"snp:75788226"), `refGene::location` = c("nonsynonymous SNV", 
"nonsynonymous SNV", "nonsynonymous SNV", "nonsynonymous SNV"
), `refGene::type` = c("ZNF717:NM_001290208:exon5:c.T2738C:p.F913S,ZNF717:NM_001128223:exon5:c.T2738C:p.F913S,ZNF717:NM_001290209:exon5:c.T2588C:p.F863S,", 
"ZNF717:NM_001290208:exon5:c.C2437T:p.P813S,ZNF717:NM_001128223:exon5:c.C2437T:p.P813S,ZNF717:NM_001290209:exon5:c.C2287T:p.P763S,", 
"ZNF717:NM_001290208:exon5:c.C2437A:p.P813T,ZNF717:NM_001128223:exon5:c.C2437A:p.P813T,ZNF717:NM_001290209:exon5:c.C2287A:p.P763T,", 
"ZNF717:NM_001290208:exon5:c.G548T:p.C183F,ZNF717:NM_001128223:exon5:c.G548T:p.C183F,ZNF717:NM_001290209:exon5:c.G398T:p.C133F,"
)), .Names = c("chr", "start", "end", "REF", "ALT", "TYPE", "refGene::location", 
"refGene::type"), row.names = c("4041", "4051", "4052", "4128"
), class = "data.frame")

Answer 1

这是一个 dplyr解决方案:

library(dplyr)
df %>% group_by(start) %>% 
       mutate(ALT = paste(ALT, collapse=",")) %>%
       distinct(start)