如何从r的数据帧内的列表中提取数字信息？

Question

I have the following type of entries in the first column of a data frame called dfgModsPepFiltered_subset :

A640-P641 = 456.123x

Trying to extract the numeric information from this with the following R script:

dfgModsPepFiltered_subset$AA <- regmatches(dfgModsPepFiltered_subset$Peptide,
        gregexpr("[[:digit:]]+", dfgModsPepFiltered_subset$Peptide))

Gives me:

c("640", "641", "453", "123")

However, what I really need is a new column for each of "640" , "641" and "456.123" .

I've tried various combinations of unlisting but can't seem to get the format right.

Answer 1

You could modify the regmatches

 as.data.frame(do.call(`rbind`,
         lapply(regmatches(dfgModsPepFiltered_subset$Peptide,
             gregexpr("[[:digit:].]+", dfgModsPepFiltered_subset$Peptide)), 
                                                        as.numeric))

  #   V1  V2      V3
  #1 640 641 456.123
  #2 620 625 285.400

Or using extract from tidyr

library(tidyr)
res <-  extract(dfgModsPepFiltered_subset, Peptide, c('Col1', 'Col2', 'Col3'),
               '[A-Z](\\d+)-[A-Z](\\d+) += +(\\d+\\.\\d+).+', convert=TRUE) 


res
#  Col1 Col2    Col3
#1  640  641 456.123
#2  620  625 285.400

Or you could use the regex

extract(dfgModsPepFiltered_subset, Peptide, c('Col1', 'Col2', 'Col3'),
        '[^0-9]+([0-9]+)[^0-9]+([0-9]+)[^0-9]+([0-9.]+)[^0-9]+')

Or

library(splitstackshape)
res1 <-  cSplit(dfgModsPepFiltered_subset, 'Peptide', '[^0-9.]', fixed=FALSE)
res1[,names(res1)[!colSums(is.na(res1))], with=FALSE]
#   Peptide_2 Peptide_4 Peptide_7
#1:       640       641   456.123
#2:       620       625   285.400

Or using strsplit

 as.data.frame(t(sapply(strsplit(dfgModsPepFiltered_subset$Peptide,
                       '[^0-9.]'), function(x) na.omit(as.numeric(x)))))

 #   V1  V2      V3
 #1 640 641 456.123
 #2 620 625 285.400

data

dfgModsPepFiltered_subset <- data.frame(Peptide= c('A640-P641 = 456.123x',
       'A620-B625 = 285.400x'), stringsAsFactors=FALSE)