R Wildcard data frame merging - regex

Question

I'm trying to merge a data frame and vector not by exact string matches in a column, but by wildcard string matches. To clarify, say you have this dataframe:
v <-data.frame(X1=c("AGTACAGT","AGTGAAGT","TGTA","GTTA","GAT","GAT"),X2=c(1,1,1,1,1,1))
# X1 X2
# 1 AGTACAGT 1
# 2 AGTGAAGT 2
# 3 TGTA 3
# 4 GTTA 4
# 5 GAT 5
# 6 GAT 6
I want to create a dataframe by creating a different color for every AGT.{3}GT,.{T|G}TA,GAT pattern, and creating a new column X3 that would show that color. So something like this:
# X1 X2 X3
# 1 AGTACAGT 1 "#FE7F01"
# 2 AGTGAAGT 2 "#FE7F01"
# 3 TGTA 3 "#FE7F00"
# 4 GTTA 4 "#FE7F00"
# 5 GAT 5 "#FE8002"
# 6 GAT 6 "#FE8002"
So far I am using this to create colors for each level, but I don't know how to count how many "wildcard levels" as opposed to singular levels there are:
x <- nlevels(v$X1)
x.colors2 <- colorRampPalette(brewer.pal(8,"Paired"))(x)
G <- data.frame("X1"=levels(v$X1),"X3"=x.colors2)
v <- merge(v,G)

Answer 1

Here's a solution.
Find patterns:
pat <- c("^AGT.{3}GT$", "^.(T|G)TA$", "^GAT$")
n <- length(pat)
indList <- lapply(pat, grep, v$X1)
Generate colors:
library(RColorBrewer)
col <- colorRampPalette(brewer.pal(8, "Paired"))(n)
Add colors to data frame:
colFull <- rep(col, sapply(indList, length))
v$color <- colFull[order(unlist(indList))]
The result:
v
# X1 X2 color
# 1 AGTACAGT 1 #A6CEE3
# 2 AGTGAAGT 1 #A6CEE3
# 3 TGTA 1 #979C62
# 4 GTTA 1 #979C62
# 5 GAT 1 #FF7F00
# 6 GAT 1 #FF7F00