-1

我有一个包含 3 列（基因、varian_type 和样本）的数据框和两列中的另一列（路径和基因）。在第二个中，我列出了每个途径中的基因。所以现在我想创建一个包含 4 列（基因、变体类型、样本和通路）的新数据框，显示每个基因存在的一个或多个通路。有谁能帮助我吗？提前致谢。

1)

     Hugo_Symbol    Variant_Type         Tumor_Sample_Barcode
1       ZAP70          SNP           TCGA-E9-A1RC-01A-11D-A159-09

2)

structure(list(circuit_names = c("hsa04014__44", "hsa04014__33", 
"hsa04014__37", "hsa04014__24", "hsa04014__26", "hsa04014__30"
), mutated = c("ZAP70,NF1,MAPK1,RAF1,CSF1R,RASGRP1,MAP2K1,MAP2K1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,NF1,PLCG1,PLCG1,PLCG1", 
"ZAP70,NF1,AKT3,CSF1R,BAD,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,PIK3R5,NF1,BCL2L1,PLCG1,PLCG1,PLCG1,AKT3", 
"ZAP70,NF1,AKT3,CSF1R,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,PIK3R5,NF1,PLCG1,PLCG1,PLCG1,FOXO4,AKT3", 
"ZAP70,NF1,CSF1R,RGL2,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,NF1,PLCG1,PLCG1,PLCG1", 
"ZAP70,NF1,CSF1R,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,NF1,PLCG1,PLCG1,PLCG1,PLCE1", 
"ZAP70,NF1,CSF1R,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,NF1,PLCG1,PLCG1,PLCG1,PLCE1"
)), row.names = c(NA, 6L), class = "data.frame")

3）我不会这样的

    structure(list(Hugo_Symbol = c("ZAP70", "TTN", "TTN", "PRKCD", 
"PIK3CA", "TLR3"), Variant_Type = c("SNP", "SNP", "SNP", "SNP", 
"SNP", "SNP"), Tumor_Sample_Barcode = c("TCGA-E9-A1RC-01A-11D-A159-09", 
"TCGA-E9-A1RC-01A-11D-A159-09", "TCGA-E9-A1RC-01A-11D-A159-09", 
"TCGA-E9-A1RC-01A-11D-A159-09", "TCGA-E9-A1RC-01A-11D-A159-09", 
"TCGA-E9-A1RC-01A-11D-A159-09"), Pathways = c("hsa04014__44, hsa04014__33, hsa04014__37, hsa04014__24", 
"hsa04530__11 20 16", "hsa04530__11 20 16", "hsa04722__37, hsa04722__35, hsa04722__33", 
"hsa04151__25, hsa04151__37, hsa04151__73", "hsa04620__23")), row.names = c("6", 
"8", "9", "11", "13", "16"), class = "data.frame")
4

1 回答 1

0

更新 -更改了解决方案方法，以便它也可以处理 OP 指出的情况。即如果Hugo_Symbol是NF1那么逻辑不应该匹配NF11或NF12

library(dplyr)
library(tidyr)

df1  %>%
  mutate(Hugo_Symbol = as.character(Hugo_Symbol)) %>%   #convert factor to character variable
  left_join(df2 %>%
              separate_rows(mutated, sep = ','), 
            by = c("Hugo_Symbol" = "mutated")) %>%
  group_by(Hugo_Symbol, Variant_Type, Tumor_Sample_Barcode) %>%
  summarise(Pathways = paste(unique(circuit_names), collapse = ","))   #combine distinct values in Pathways

这使

  Hugo_Symbol Variant_Type Tumor_Sample_Barcode         Pathways                                     
1 NF1         SNP          TCGA-E9-A1RC-01A-11D-A159-09 hsa04014__44,hsa04014__33


样本数据：

df1 <- structure(list(Hugo_Symbol = "NF1", Variant_Type = "SNP", Tumor_Sample_Barcode = "TCGA-E9-A1RC-01A-11D-A159-09"), .Names = c("Hugo_Symbol", 
"Variant_Type", "Tumor_Sample_Barcode"), class = "data.frame", row.names = "1")

df2 <- structure(list(circuit_names = c("hsa04014__44", "hsa04014__33", 
"hsa04014__37", "hsa04014__24"), mutated = c("ZAP70,NF1,MAPK1,RAF1,CSF1R,RASGRP1,MAP2K1", 
"ZAP70,NF1,AKT3,CSF1R,BAD,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1", 
"ZAP70,NF11,AKT3,CSF1R,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1,RASGRF", 
"ZAP70,NF12,CSF1R,RGL2,RASGRP1,RASGRF1,RASGRF1,RASGRF1,RASGRF1"
)), .Names = c("circuit_names", "mutated"), class = "data.frame", row.names = c("1", 
"2", "3", "4"))
于 2018-07-05T07:47:45.520 回答