如何在字符向量列表上使用 seqinr::computePI

Question

This is what the input data looks like (representative sample).

sample1        MAQSVNIQDQYLNQ
sample2        MAADRAQNLQDTFLNHV
sample3        MAERSQNLQ

I am trying to use the computePI() function from the seqinr library to perform a calculation on the second column. To do this, I need to convert the character strings in the second column to a vector of single characters, and use that as input for computePI() . To do this, I tried using strsplit() . It works on an individual string:

library(seqinr)
str <- c("MAQSVNIQDQYLNQ")
unlist(strsplit(str, split = ""))
#R> [1] "M" "A" "Q" "S" "V" "N" "I" "Q" "D" "Q" "Y" "L" "N" "Q"

Followed by:

computePI(unlist(strsplit(str, split = "")))
#R> [1] 3.799404

After saving the input data read using read.csv to a variable:

data <- read.csv("filename", header = FALSE, sep = "\t")

I tried the following:

computePI(unlist(strsplit(data$V2, split = "")))

But got an error telling me that the input isn't a character string. How can I fix this?

Answer 1

您可以使用sapply ：

sapply(strsplit(data$V2, split = ""), seqinr::computePI)

Answer 2

You can use Vectorize :

library(seqinr)
VcomputePI <- Vectorize(computePI)
VcomputePI(strsplit(dat$V2, split = ""))
#R> [1] 3.799404 5.194135 5.753213

Or sapply / lapply / vapply like as Ronak Shah shows. vapply is likely the fastest:

vapply(strsplit(dat$V2, split = ""), computePI, 0.)
#R> [1] 3.799404 5.194135 5.753213

Data

dat <- data.frame(
  V1 = paste0("sample", 1:3), 
  V2 = c("MAQSVNIQDQYLNQ", "MAADRAQNLQDTFLNHV", "MAERSQNLQ"), 
  stringsAsFactors = FALSE)