Hacer coincidir una columna de un marco de datos con las columnas de otro marco de datos y, si coinciden, agregar una nueva columna

Question

Hacer coincidir una columna de un marco de datos con las columnas de otro marco de datos y, si coinciden, agregar una nueva columna

I got two big data frames, one (df1) has this structure

    V1    V2    V3
1  Chr1  7507 10944
2  Chr1 10944 13170
3  Chr1 13170 20065
4  Chr1 20065 28273
5  Chr1 28273 29960
6  Chr1 29960 36599
7  Chr1 36599 37513
8  Chr1 37513 40360
9  Chr1 40360 48796
10 Chr1 48796 50661

The other (df2) has this

     V1    V2    V3 V4  V5
1  Chr1  7507  7507  1   1
2  Chr1 10944 10944  1   2
3  Chr1 13170 13170  1  22
4  Chr1 20065 20065  1   3
5  Chr1 28273 28273  1 161
6  Chr1 29960 29960  1  10
7  Chr1 36599 36599  1 604
8  Chr1 37513 37513  1 117
9  Chr1 40360 40360  1   8
10 Chr1 48796 48796  1   3

what I'm trying to do is to check if the column V2 or V3 (is the same) of df2 is = or between the range of V2 and V3 of df1 then I want to write the value of V5 of df2 in a new column in df1 if not write 0. the result that i want would be like :

Chr1    7507    10944   1
Chr1    10944   13170   2   
Chr1    13170   20065   22  
Chr1    20065   28273   3   
Chr1    28273   29960   161 
Chr1    29960   36599   10  
Chr1    36599   37513   604 
Chr1    37513   40360   117 
Chr1    40360   48796   8
.
.
.

Do you know any good way to do this? Thank you very much.

preguntado el 28 de mayo de 14 a las 14:05

user3683485

Creo que necesitas usar merge. You can check the example in R help files (?merge) - dickoa

In your example, all V2 and V3 values of df2 have an exact match in V2 of df1. If this applies to your whole data, then a relatively simple merge is appropriate as suggested by @dickoa. If your actual data is different (so that you would need to check ranges), it would be better if you could also edit your sample data - talat

1 Respuestas

No es la respuesta que estás buscando? Examinar otras preguntas etiquetadas r dataframe multiple-columns matching or haz tu propia pregunta.

Creo que necesitas usar merge. You can check the example in R help files (?merge) - dickoa
In your example, all V2 and V3 values of df2 have an exact match in V2 of df1. If this applies to your whole data, then a relatively simple merge is appropriate as suggested by @dickoa. If your actual data is different (so that you would need to check ranges), it would be better if you could also edit your sample data - talat

Jaap · Accepted Answer

As @beginneR already mentioned in the comments, all V2 y V3 valores de df2 have an exact match with V2 of df1. If I interpret your question correctly, this is probably not what you wanted. The following example is what I yhink you are looking for.

Reading the two dataframes:

df1 <- read.table(header=TRUE, text="rn    V1    V2    V3
1  Chr1  7507 10944
2  Chr1 10944 13170
3  Chr1 13170 20065
4  Chr1 20065 28273
5  Chr1 28273 29960
6  Chr1 29960 36599
7  Chr1 36599 37513
8  Chr1 37513 40360
9  Chr1 40360 48796
10 Chr1 48796 50661")

df2 <- read.table(header=TRUE, text="rn     V1    V2    V3 V4  V5
1  Chr1  7507  7507  1   1
2  Chr1 10944 10944  1   2
3  Chr1 13170 13170  1  22
4  Chr1 20065 20065  1   3
5  Chr1 28273 28273  1 161
6  Chr1 29960 29960  1  10
7  Chr1 36599 36599  1 604
8  Chr1 37513 37513  1 117
9  Chr1 40360 40360  1   8
10 Chr1 48796 48796  1   3")

Getting rid of V3 in df2 as it is exactly the same as V2:

df2 <- df2[,-4]

Making the values in V2 of df2 higher

df2$V2 <- df2$V2 + 2000

Con la ifelse function you can assign the values of V5 to a new variable in df1 when the meet the requirements:

df1$V4 <- ifelse(df2$V2 >= df1$V2 & df2$V2 <= df1$V3, df2$V5, 0)