提取与模式匹配的特定单词

Question

I have data frame with a column:

nf1$Info = AC=1;AF=0.500;AN=2;BaseQRankSum=-1.026e+00;ClippingRankSum=-1.026e+00;DP=4;ExcessHet=3.0103;FS=0.000;MLEAC=1;MLEAF=0.500;MQ=28.25;MQRankSum=-1.026e+00;QD=10.18;ReadPosRankSum=1.03;SOR=0.693

I'm trying to extract a specific value from this column.

For eg I'm interested in "MQRankSum" and I used:

str_extract(nf1$Info,"[MQRankSum]+=[:punct:]+[0-9]+[.]+[0-9]+")

It returns value for BaseQRankSum instead of MQRankSum .

Answer 1

Including characters into square brackets creates a character class matching any of the defined characters, so [yes]+ matches yyyyyyyyy , eyyyyss , etc.

What you want to do is to match a word MQRankSum , = , and then any chars other than ; :

str_extract(nf1$Info,"MQRankSum=[^;]+")

If you want to exlcude MQRankSum= from the match, use a lookbehind:

str_extract(nf1$Info,"(?<=MQRankSum=)[^;]+")
                      ^^^^^^^^^^^^^^^

The (?<=MQRankSum=) positive lookbehind will make sure there is MQRankSum= text immediately to the left of the current location, and only after that will match 1 or more chars other than ; .

Answer 2

We could split INFO column into multiple columns then extract desired column:

# dummy data
df1 <- data.frame(x = 1:3,
                  info = c("AC=1;AF=0.500;MQRankSum=2;BaseQRankSum=-1.026e+00;ClippingRankSum=-1.026e+00;",
                           "AC=1;AF=0.500;MQRankSum=2;ClippingRankSum=-1.026e+00;DP=4;",
                           "AN=2;BaseQRankSum=-1.026e+00;"),
                  stringsAsFactors = FALSE)

# split INFO into seperate columns
df1_info <- data.table::rbindlist(
  lapply(strsplit(df1$info, ";|="), function(i)
    setNames(data.frame(t(as.numeric(i[ c(FALSE, TRUE) ]))), i[ c(TRUE, FALSE) ])
    ),
  fill = TRUE)

df1_info
#    AC  AF MQRankSum BaseQRankSum ClippingRankSum DP AN
# 1:  1 0.5         2       -1.026          -1.026 NA NA
# 2:  1 0.5         2           NA          -1.026  4 NA
# 3: NA  NA        NA       -1.026              NA NA  2

# extract required column 
df1_info$BaseQRankSum
# [1] -1.026     NA -1.026

---

VCF INFO standard :

Various site-level annotations. The annotations contained in the INFO field are represented as tag-value pairs, where the tag and value are separated by an equal sign, ie =, and pairs are separated by colons, ie ; as in this example:
MQ=99.00;MQ0=0;QD=17.94 .