CODEWITHSUNDEEP
rdataframe
abraham
abraham

Reputation: 743

R construct a data frame using 2 column

I have a data frame like:

df

    group group_name      value
1       1       <NA>     VV0001
2       1       <NA> VV_RS00280
3       2       <NA>     VV0002
4       2       <NA> VV_RS00285
5       3       <NA>     VV0003
6       3       <NA> VV_RS00290
7       5       <NA>     VV0004
8       5       <NA> VV_RS00295
9       6       <NA>     VV0005
10      6       <NA> VV_RS00300
11      7       <NA>     VV0006
12      7       <NA> VV_RS00305
13      8       <NA>     VV0007
14      8       <NA> VV_RS00310
15      9       <NA>     VV0009
16      9       <NA> VV_RS00315
17     10       <NA>     VV0011
18     10       <NA> VV_RS00320
19     11       <NA>     VV0012
20     11       <NA> VV_RS00325
21     12       <NA>     VV0013
22     12       <NA> VV_RS00330

so I want to construct an other data frame using the columns "group" and "value", all the group 1 (df[df$group == 1,]) will get the data in "value" column (VV0001, VV_RS00280) and construct the data.frame like:

   group  value
1  VV0001 VV_RS00280

and then the next df[df$group == 2,], and so on, at the end will be:

   group  value
1  VV0001 VV_RS00280
2  VV0002 VV_RS00285
3  VV0003 VV_RS00290
4  VV0004 VV_RS00295

I tried to do it manually but the nrow(df) is big, > 3000 or more !!

Thanks

Upvotes: 1

Views: 47

Answers (1)

Kra.P
Kra.P

Reputation: 15123

You may try,

library(dplyr)
library(tidyr)

df %>%
  rename(idv = group) %>%
  mutate(group_name = rep(c("group", "value"),n()/2)) %>%
  group_by(idv) %>%
  pivot_wider(names_from = group_name, values_from = value) %>%
  ungroup %>%
  select(-idv)

   group  value     
   <chr>  <chr>     
 1 VV0001 VV_RS00280
 2 VV0002 VV_RS00285
 3 VV0003 VV_RS00290
 4 VV0004 VV_RS00295
 5 VV0005 VV_RS00300
 6 VV0006 VV_RS00305
 7 VV0007 VV_RS00310
 8 VV0009 VV_RS00315
 9 VV0011 VV_RS00320
10 VV0012 VV_RS00325
11 VV0013 VV_RS00330

Upvotes: 4

Related Questions