Separates locus tags mentioned in full text and expands ranges like YPO1970-74 into new rows
Examples
x <- data.frame(row = 1, text = "some genes like YPO1002 and YPO1970-74")
separate_tags(x, "YPO")
#> id match row text
#> 1 YPO1002 YPO1002 1 some genes like YPO1002 and YPO1970-74
#> 1.1 YPO1970 YPO1970-74 1 some genes like YPO1002 and YPO1970-74
#> 1.1.1 YPO1971 YPO1970-74 1 some genes like YPO1002 and YPO1970-74
#> 1.1.2 YPO1972 YPO1970-74 1 some genes like YPO1002 and YPO1970-74
#> 1.1.3 YPO1973 YPO1970-74 1 some genes like YPO1002 and YPO1970-74
#> 1.1.4 YPO1974 YPO1970-74 1 some genes like YPO1002 and YPO1970-74