topic Re: Using arrays and substr function for sequencing DNA in SAS Procedures

Using arrays and substr function for sequencing DNA

zjhansen30 — Mon, 22 Feb 2016 22:30:13 GMT

I am new at arrasy, but I am trying to get the below 15 sequences of DNA, which are 60 characters, to be 60 variables, D1-D60, where D1 holds the first position, D2 the second position, and so on.  I have never used the substr function, so I am probably doing it incorrectly. Any help would be great

data dna2 (drop=dna i);
set dna;

array d(60);
do i=1 to 60;
	d(i)=d(4/(i));
	dna=substr (dna, 1, 15);
end;
run;

data dna;
	length dna $ 60;
	input dna $;
datalines;
TGGAAGGGCTAATTTGGTCCCAAAAAAGACAAGAGATCCTTGATCTGTGGATCTACCACA
TGATTGGCAGAACTACACACCAGGGCCAGGGATCAGATATCCACTGACCTTTGGATGGTG
CTTCAAGTTAGTACCAGTTGAACCAGAGCAAGTAGAAGAGGCCAAATAAGGAGAGAAGAA
CAGCTTGTTACACCCTATGAGCCAGCATGGGATGGAGGACCCGGAGGGAGAAGTATTAGT
GTGGAAGTTTGACAGCCTCCTAGCATTTCGTCACATGGCCCGAGAGCTGCATCCGGAGTA
CTACAAAGACTGCTGACATCGAGCTTTCTACAAGGGACTTTCCGCTGGGGACTTTCCAGG
GAGGTGTGGCCTGGGCGGGACTGGGGAGTGGCGAGCCCTCAGATGCTACATATAAGCAGC
TGCTTTTTGCCTGTACTGGGTCTCTCTGGTTAGACCAGATCTGAGCCTGGGAGCTCTCTG
GCTAACTAGGGAACCCACTGCTTAAGCCTCAATAAAGCTTGCCTTGAGTGCTCAAAGTAG
TGTGTGCCCGTCTGTTGTGTGACTCTGGTAACTAGAGATCCCTCAGACCCTTTTAGTCAG
TGTGGAAAATCTCTAGCAGTGGCGCCCGAACAGGGACTTGAAAGCGAAAGTAAAGCCAGA
GGAGATCTCTCGACGCAGGACTCGGCTTGCTGAAGCGCGCACGGCAAGAGGCGAGGGGCG
GCGACTGGTGAGTACGCCAAAAATTTTGACTAGCGGAGGCTAGAAGGAGAGAGATGGGTG
CGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGGAAAAAATTCGGTTAAGGC
CAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCAGGGAGCTAGAAC
;

Re: Using arrays and substr function for sequencing DNA

FreelanceReinh — Mon, 22 Feb 2016 22:40:14 GMT

How about this?

data dna2 (drop=dna i);
set dna;
array d(60) $1;
do i=1 to 60;
  d(i)=char(dna,i);
end;
run;

Edit: Alternatively, you could use the SUBSTR function:

d(i)=substr(dna,i,1);

Re: Using arrays and substr function for sequencing DNA

zjhansen30 — Mon, 22 Feb 2016 22:39:30 GMT

That works, I almost had it, but do you know how to use the substr function doing it at all?

Re: Using arrays and substr function for sequencing DNA

FreelanceReinh — Mon, 22 Feb 2016 22:44:24 GMT

Please see the edited post.

The CHAR function syntax is a bit shorter. With SUBSTR you have to specify where to start the substring (second argument) and how many characters it should contain (third argument). The latter is implied to be 1 with the CHAR function.

Re: Using arrays and substr function for sequencing DNA

zjhansen30 — Mon, 22 Feb 2016 22:48:01 GMT

Thank you, that makes sense to use the char over the substr.....and thank you for your help

Re: Using arrays and substr function for sequencing DNA

Haikuo — Tue, 23 Feb 2016 02:41:33 GMT

This maybe a little obscured for you, but depending on the size of your data set, it will save you some time considerably.

data want;
set dna;
array d(60) $1;
call pokelong(dna, addrlong(d(1)),60);
drop dna;
run;

Re: Using arrays and substr function for sequencing DNA

FreelanceReinh — Tue, 23 Feb 2016 09:48:37 GMT

@Haikuo: Thanks for pointing out this interesting alternative. I had never used this call routine. The warnings in the documentation are quite intimidating, though ("devastating problems ... destroying a vital element ..."). So, maybe a bit too advanced for someone who had "never used the substr function."

Re: Using arrays and substr function for sequencing DNA

Kurt_Bremser — Tue, 23 Feb 2016 10:13:48 GMT

You could read the data directly into the array like that:

data dna;
array dna {60} $1 dna1-dna60;
do i = 1 to 60;
input dna{i} $1.@;
end;
drop i;
datalines;
TGGAAGGGCTAATTTGGTCCCAAAAAAGACAAGAGATCCTTGATCTGTGGATCTACCACA
TGATTGGCAGAACTACACACCAGGGCCAGGGATCAGATATCCACTGACCTTTGGATGGTG
CTTCAAGTTAGTACCAGTTGAACCAGAGCAAGTAGAAGAGGCCAAATAAGGAGAGAAGAA
CAGCTTGTTACACCCTATGAGCCAGCATGGGATGGAGGACCCGGAGGGAGAAGTATTAGT
GTGGAAGTTTGACAGCCTCCTAGCATTTCGTCACATGGCCCGAGAGCTGCATCCGGAGTA
CTACAAAGACTGCTGACATCGAGCTTTCTACAAGGGACTTTCCGCTGGGGACTTTCCAGG
GAGGTGTGGCCTGGGCGGGACTGGGGAGTGGCGAGCCCTCAGATGCTACATATAAGCAGC
TGCTTTTTGCCTGTACTGGGTCTCTCTGGTTAGACCAGATCTGAGCCTGGGAGCTCTCTG
GCTAACTAGGGAACCCACTGCTTAAGCCTCAATAAAGCTTGCCTTGAGTGCTCAAAGTAG
TGTGTGCCCGTCTGTTGTGTGACTCTGGTAACTAGAGATCCCTCAGACCCTTTTAGTCAG
TGTGGAAAATCTCTAGCAGTGGCGCCCGAACAGGGACTTGAAAGCGAAAGTAAAGCCAGA
GGAGATCTCTCGACGCAGGACTCGGCTTGCTGAAGCGCGCACGGCAAGAGGCGAGGGGCG
GCGACTGGTGAGTACGCCAAAAATTTTGACTAGCGGAGGCTAGAAGGAGAGAGATGGGTG
CGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGGAAAAAATTCGGTTAAGGC
CAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCAGGGAGCTAGAAC
;
run;

Re: Using arrays and substr function for sequencing DNA

Haikuo — Tue, 23 Feb 2016 13:54:21 GMT

@FreelanceReinh, True and agreed. Direct memory write-operation has inherited risk. We implemented a few in the case where huge amount of data manipulation is required, and it does help. Here is just to add some new elements to the discussion, after all, besides seeking answers for specific questions, people also want to learn.