CODEWITHSUNDEEP
r
Nikolas Service
Nikolas Service

Reputation: 69

How can I identify the rows based on one string in a sentence

I recently asked a question which was very useful and I tried to use the same approach to find my solution 

df<- structure(list(How = structure(c(2L, 2L, 2L, 1L, 2L, 2L, 2L, 
1L, 2L, 2L, 2L, 1L, 1L, 2L, 2L, 2L, 2L, 1L, 1L, 2L), .Label = c("Ismainbody", 
"IsmainbodyCandidate"), class = "factor"), No = c(12L, 38L, 38L, 
3L, 49L, 38L, 85L, 4L, 38L, 57L, 38L, 5L, 6L, 10L, 4L, 12L, 38L, 
7L, 8L, 61L), Main = structure(c(6L, 13L, 9L, 15L, 20L, 12L, 
1L, 19L, 10L, 2L, 7L, 18L, 4L, 14L, 5L, 16L, 8L, 3L, 17L, 11L
), .Label = c("Daa_ASTRONOMY Iso 1B of Tn-1 TTT=ASTY ", "E7EUT5_ASTRONOMY gas TTT=ASTY BOO=3 ", 
"ECO", "ECO transferase E [TTT=ASTY]", "ECO_ASTRONOMY karim,  TTT=ASTY BOO=3", 
"FSSZ1_ASTRONOMY Karim, tyBOO II brothers 74 TTT=ASTY BOO=3 ", 
"H2A1A_ASTRONOMY  tyBOO 1-A TTT=ASTY BOO=1 ", "H2A2B_ASTRONOMY tyBOO 2-B TTT=ASTY BOO=1 ", 
"H2A3_ASTRONOMY Hammer H2A tyBOO 3 TTT=ASTY BOO=1 ", "H2AV_ASTRONOMY Iso 2 of  TTT=ASTY ", 
"H2E_ASTRONOMY ufidm TTT=ASTY ", "Hammer [TTT=ASTY]", "Hammer H2A tyBOO 2-C [TTT=ASTY]", 
"Iso 2 of Deleted in house [TTT=ASTY]", "Iso 2019 of denis [TTT=ASTY]", 
"K2C74_ASTRONOMY karim, tyBOO II  TTT=ASTY BOO=1", "KAR_ASTRONOMY karim, tyBOO TTT=ASTY BOO=1 BBS", 
"karim, tyBOO II  1b [TTT=ASTY]", "karim, tyBOO II 7 [TTT=ASTY]", 
"Putative heat 7 [TTT=ASTY]"), class = "factor")), class = "data.frame", row.names = c(NA, 
-20L))

This is the the data I have and I want to remove those rows that have the following letters in them:

karim

ECO

Daa

I did like this 

lookm <- c("karim", "ECO", "Daa") 
df2<- df[!df$Main %in% lookm, ]

but nothing happened. How can I do this?

Upvotes: 4

Views: 92

Answers (3)

allanvc
allanvc

Reputation: 1156

We could use the str_detect() function from package stringr:

library(stringr)
df[!str_detect(df$Main, "karim|ECO|Daa"),]

output:

                   How No                                                        Main
1  IsmainbodyCandidate 12 FSSZ1_ASTRONOMY Karim, tyBOO II brothers 74 TTT=ASTY BOO=3 
2  IsmainbodyCandidate 38                             Hammer H2A tyBOO 2-C [TTT=ASTY]
3  IsmainbodyCandidate 38           H2A3_ASTRONOMY Hammer H2A tyBOO 3 TTT=ASTY BOO=1 
4           Ismainbody  3                                Iso 2019 of denis [TTT=ASTY]
5  IsmainbodyCandidate 49                                  Putative heat 7 [TTT=ASTY]
6  IsmainbodyCandidate 38                                           Hammer [TTT=ASTY]
9  IsmainbodyCandidate 38                          H2AV_ASTRONOMY Iso 2 of  TTT=ASTY 
10 IsmainbodyCandidate 57                        E7EUT5_ASTRONOMY gas TTT=ASTY BOO=3 
11 IsmainbodyCandidate 38                  H2A1A_ASTRONOMY  tyBOO 1-A TTT=ASTY BOO=1 
14 IsmainbodyCandidate 10                        Iso 2 of Deleted in house [TTT=ASTY]
17 IsmainbodyCandidate 38                   H2A2B_ASTRONOMY tyBOO 2-B TTT=ASTY BOO=1 
20 IsmainbodyCandidate 61                               H2E_ASTRONOMY ufidm TTT=ASTY

If you also want to identify the word "Karim" with the uppercase "K", you can try:

df[!str_detect(df$Main, "(K|k)arim|ECO|Daa"),]

output:

                   How No                                              Main
2  IsmainbodyCandidate 38                   Hammer H2A tyBOO 2-C [TTT=ASTY]
3  IsmainbodyCandidate 38 H2A3_ASTRONOMY Hammer H2A tyBOO 3 TTT=ASTY BOO=1 
4           Ismainbody  3                      Iso 2019 of denis [TTT=ASTY]
5  IsmainbodyCandidate 49                        Putative heat 7 [TTT=ASTY]
6  IsmainbodyCandidate 38                                 Hammer [TTT=ASTY]
9  IsmainbodyCandidate 38                H2AV_ASTRONOMY Iso 2 of  TTT=ASTY 
10 IsmainbodyCandidate 57              E7EUT5_ASTRONOMY gas TTT=ASTY BOO=3 
11 IsmainbodyCandidate 38        H2A1A_ASTRONOMY  tyBOO 1-A TTT=ASTY BOO=1 
14 IsmainbodyCandidate 10              Iso 2 of Deleted in house [TTT=ASTY]
17 IsmainbodyCandidate 38         H2A2B_ASTRONOMY tyBOO 2-B TTT=ASTY BOO=1 
20 IsmainbodyCandidate 61                     H2E_ASTRONOMY ufidm TTT=ASTY

Upvotes: 2

allanvc
allanvc

Reputation: 1156

We can think of a tidyverse approach combining dplyr's slice() and stringr's str_which() functions:

library(dplyr)
library(stringr)

df %>%
  slice(-str_which(df$Main, "karim|ECO|Daa"))

#OR
#df %>%
#  slice(-str_which(df$Main, "(K|k)karim|ECO|Daa"))

Notice in the output below that we lost the former row indices of our original data.frame: 

                   How No                                                        Main
1  IsmainbodyCandidate 12 FSSZ1_ASTRONOMY Karim, tyBOO II brothers 74 TTT=ASTY BOO=3 
2  IsmainbodyCandidate 38                             Hammer H2A tyBOO 2-C [TTT=ASTY]
3  IsmainbodyCandidate 38           H2A3_ASTRONOMY Hammer H2A tyBOO 3 TTT=ASTY BOO=1 
4           Ismainbody  3                                Iso 2019 of denis [TTT=ASTY]
5  IsmainbodyCandidate 49                                  Putative heat 7 [TTT=ASTY]
6  IsmainbodyCandidate 38                                           Hammer [TTT=ASTY]
7  IsmainbodyCandidate 38                          H2AV_ASTRONOMY Iso 2 of  TTT=ASTY 
8  IsmainbodyCandidate 57                        E7EUT5_ASTRONOMY gas TTT=ASTY BOO=3 
9  IsmainbodyCandidate 38                  H2A1A_ASTRONOMY  tyBOO 1-A TTT=ASTY BOO=1 
10 IsmainbodyCandidate 10                        Iso 2 of Deleted in house [TTT=ASTY]
11 IsmainbodyCandidate 38                   H2A2B_ASTRONOMY tyBOO 2-B TTT=ASTY BOO=1 
12 IsmainbodyCandidate 61                               H2E_ASTRONOMY ufidm TTT=ASTY

Upvotes: 1

akrun
akrun

Reputation: 887651

We could use grep

df[!grepl(paste(lookm, collapse="|"), df$Main),]

Upvotes: 3

Related Questions