regex - How to search and extract in a dataframe in #R using a pattern that includes moving from columns and rows?

Question

I would like to search and extract from a big dataframe in R a number that is located in different places but is always located one column and one row more than the word "LAYER".

I have tried to do it but I can only find a way to extract "LAYER" with a filter using the dplyr package but maybe I need to use something different.

For example, in the following table I would like to extract the 2 and the 3.65 that are bold in the dataframe.

X1	X2	X3
LAYER	1	NA
190	2	NA
NA	20	1200
NA	30	2200
...	...	...
LAYER	2	NA
180	3.65	NA
NA	50	1850
NA	95	2300
...	...	...

Answer 1

You can get the indices where X1 is 'LAYER' and extract next X2 value.

inds <- which(data$X1 == 'LAYER')
result <- data.frame(X2 = data$X2[inds + 1])