Merge by Range in R - Applying Loops

Question

I posted a question here: Matched Range Merge in R about merging two files based on a number in one file falling into a range in the second file. Thus far, I have been unsuc

Answer 1

I believe what you're asking for is a conditional join. They're easy in SQL, and the sqldf package makes it easy to query data frames in R using SQL.

Just pick a version depending on how you want unmatched SNPs handled.

Inner join version:

> sqldf("select * from file1test f1 inner join file2 f2 
+       on (f1.BP > f2.BP_start and f1.BP<= f2.BP_end) ")

Output:

     SNP     BP  Gene BP_start BP_end
1 rs2343 860269 E3543   860260 879955
2  rs754 861822 E3543   860260 879955
3  rs754 861822   E11   861322 879533
4  rs854 367934  E613   367640 368634
> 

Left Join version:

> sqldf("select * from file1test f1 left join file2 f2 
+       on (f1.BP > f2.BP_start and f1.BP<= f2.BP_end) ")

Output:

     SNP     BP  Gene BP_start BP_end
1 rs2343 860269 E3543   860260 879955
2  rs211 369640         NA     NA
3  rs754 861822 E3543   860260 879955
4  rs754 861822   E11   861322 879533
5  rs854 367934  E613   367640 368634
6  rs343 706940         NA     NA
7  rs626 717244         NA     NA
> 

Note that you may want to be careful where you place the = if it matters which group a BP will fall in for the case where a BP exactly matches a BP_start or BP_end.