CODEWITHSUNDEEP
r
Phantom Photon
Phantom Photon

Reputation: 808

Creating a new variable column based on data from another column

I'm pretty new to R, and programming in general, and I'm wondering the best way to loop through a column so I can add a column to the data frame further describing the observations I looped through.

I currently have a list of amino acids and their positions on a protein that looks like this:

Residue Position
H   1
R   2
K   3
D   4
E   5
H   6
R   7
K   8
D   9
E   10

I'd like something that looks like this (where H, R, and K are basic amino acids, and D and E are acidic amino acids):

Residue Position    Properties
H   1   Basic
R   2   Basic
K   3   Basic
D   4   Acidic
E   5   Acidic
H   6   Basic
R   7   Basic
K   8   Basic
D   9   Acidic
E   10  Acidic

I'm really not sure where to start, and I'm having difficulty finding a good resource for this kind of situation in R.

I started by trying to subset the data, but then I realized that wouldn't do the trick:

Basic
h.dat <- subset(all, all$Residue == "H")
r.dat <- subset(all, all$Residue == "R")
k.dat <- subset(all, all$Residue == "K")

Acidic
d.dat <- subset(all, all$Residue == "D")
e.dat <- subset(all, all$Residue == "E")

Thanks!

Note: 
H = Histidine (Basic amino acid)
R = Arginine (Basic)
K = Lysine (Basic)

E = Glutamic Acid (Acidic)
D = Aspartic Acid (Acidic)

Upvotes: 1

Views: 2438

Answers (3)

Rich Scriven
Rich Scriven

Reputation: 99371

You can use ifelse. If df is the name of your original data,

df$Property <- ifelse(df$Residue %in% c("H", "R", "K"), "Basic", "Acidic")
df
#    Residue Position Property
# 1        H        1    Basic
# 2        R        2    Basic
# 3        K        3    Basic
# 4        D        4   Acidic
# 5        E        5   Acidic
# 6        H        6    Basic
# 7        R        7    Basic
# 8        K        8    Basic
# 9        D        9   Acidic
# 10       E       10   Acidic

Upvotes: 6

IRTFM
IRTFM

Reputation: 263471

I think you might want to allow for non-polar amino acids as well:

c(rep("Basic",3),rep("Acidic",2),"Non-Polar")[   # those are the choices
        match(dat$Residue, c("H","R","K","E","D"), nomatch=6) ] #select indices

So I added an 11th residue named "Z" and tested:

> dat$Property <- c(rep("Basic",3),rep("Acidic",2),"Non-Polar")[
                 match(dat$Residue, c("H","R","K","E","D"), nomatch=6) ]
> dat
   Residue Position  Property
1        H        1     Basic
2        R        2     Basic
3        K        3     Basic
4        D        4    Acidic
5        E        5    Acidic
6        H        6     Basic
7        R        7     Basic
8        K        8     Basic
9        D        9    Acidic
10       E       10    Acidic
11       Z       11 Non-Polar

Upvotes: 2

rnso
rnso

Reputation: 24623

Try:

> df1
   Residue Position
1        H        1
2        R        2
3        K        3
4        D        4
5        E        5
6        H        6
7        R        7
8        K        8
9        D        9
10       E       10

Create a reference table:

> df2
  Residue Property
1       H    Basic
2       R    Basic
3       K    Basic
4       D   Acidic
5       E   Acidic

Then merge:

> merge(df1, df2)
   Residue Position Property
1        D        9   Acidic
2        D        4   Acidic
3        E        5   Acidic
4        E       10   Acidic
5        H        1    Basic
6        H        6    Basic
7        K        8    Basic
8        K        3    Basic
9        R        7    Basic
10       R        2    Basic

Upvotes: 2

Related Questions