1
$\begingroup$

I have two files like

> head(a)
  chr position
1   1   136962
2   1   562020
3   1   672948
> class(a)
[1] "data.frame"
>

And

> dim(b)
[1] 1855235      10
> head(b[,1:3])
            SNP chr position
1:  rs62635286          1    13116
2:  rs75454623          1    14930
3:    rs806731          1    30923
4: rs200943160          1    49298
5: rs116400033          1    51479
6: rs141149254          1    54490
> class(b)
[1] "data.table" "data.frame"
>

I want to find corresponding SNP in file b matched to position in file a so I did like this that gives memory allocation error. 

> unique(merge(a,b,x.by=a$position,y.by=b$position)) 
Error: cannot allocate vector of size 137.2 Gb

I also tried rstudio cloud but uploading the bigger file never completes

Can you help me with task please?

The expected output would be something like

rs62635286  1   13116
rs376723915 1   91515
rs147061536 1   92858
rs62642131  1   135982
rs371474651 1   158006
rs201293782 1   665401
rs138476838 1   668374
rs1401137   1   2063094
rs3128293   1   2065339

EDITED

> setDT(a)
> setkey(a, chr, position)
> setkey(b, chr, position)
Error in setkeyv(x, cols, verbose = verbose, physical = physical) : 
  some columns are not in the data.table: chr,position
> setkey(b, Chromosome, Position)
> a[b, nomatch=0]
Error in bmerge(i, x, leftcols, rightcols, roll, rollends, nomatch, mult,  : 
  Incompatible join types: x.chr (factor) and i.Chromosome (integer). Factor columns must join to factor or character columns.
>
Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

You could try something like this :

First you should work with the same class of object : data.table for fast merging operation

# Turning a to a data.table
setDT(a)
# b is already a data.table

Then you might have to set the keys of both table : 

# I added chr in case you need to match the same position on a different chromosome.
setkey(a, chr, position)
setkey(b, chr, position)

Finally perform the joining operation :

a[b, nomatch=0]

Because you didn't provide the expected output, it might need some tuning !

Improve this answer
$\endgroup$
2
  • $\begingroup$ Thank you I added the output of your command in my main post. I obtained error please can you have a look? $\endgroup$
    – Angel
    Dec 18, 2019 at 16:42
  • $\begingroup$ This error means the type of your chromosome columns don't match. You can check the type of all columns with str(a). I suggest you keep the the integer format of the b table. This should update a column from factor to integer : a[, chr := as.integer(chr)]. It may work then ! $\endgroup$
    – Paul Endymion
    Dec 18, 2019 at 18:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.