Can A Snp Have A Different Effect On Different Splice Variants?

Question

[Andrea_Bio]

Hi

I recently asked a question as to whether exons can have different reading frames in the same gene. The answer was, surprisingly to me, yes. In this question I am assuming for simplicity that an exon has the same reading frame in all transcripts it features in.

Lets imagine a gene running from bases 1 to 200. Exon 1 is from bases 50 to 100 and exon 2 is from 125-150 and exon 3 is from 175-200.

Then lets say there is a splice variant (variant 1) containing exons 1 and 2 and another splice variant (variant 2) containing exons 1 and 3.

Then lets imagine there is a SNP in exon 1 at base 60. Could this SNP have a different effect in the different transcripts? For example is it possible that in variant 1 the exon is not translated at base 60 so the SNP is an an upstream untranslated region, but in variant 2 the SNP is in the translated region and causes a synonymous (or non-syn) amino acid change?

So another way of asking the question is this: if an exon appears in multiple transcripts, does it always play the same role in those transcripts? E.g is it always translated or not translated, and if it is translated, does translation start at the same place. I am assuming the answer to this will be 'no' because if an exon can have multiple reading frames in the same gene then all sorts of other scenarios are possible.

If what I describe does happen, roughly how common or uncommon is it?

Many thanks in advance

[Khader Shameer]

Hope this recent paper from Butte Lab could be interesting to you: Non-Synonymous and Synonymous Coding SNPs Show Similar Likelihood and Effect Size of Human Disease Association

Answer 1

[Pierre Lindenbaum]

I've piped the program I wrote for your previous question and described here to cross the data with dbSNP/UCSC:

java BioStar3034  | head 1000 |  tr ":-" "," |\
awk -F ',' 'BEGIN { printf("select * from snp130 where chrom=\"_\" ");} {printf(" OR (chrom=\"%s\" and chromStart=%d)\n",$1,int($2)-1);} END {printf(";\n");}' |\
mysql  -h  genome-mysql.cse.ucsc.edu -A -u genome -D hg18

One of the result is rs1152725 http://genome.ucsc.edu/cgi-bin/hgTracks?db=hg18&position=chr1%3A1209324-1209324

If the alternate base is 'C' instead of ref='T'

• transcript 1: codon CTC (L) -> CCC (Proline) (non-synonymous)
• transcript 2: codon TCT(Serine) -> TCC (Serine) (synonymous)

so my answer is YES: it happens :-)

[Khader Shameer]

Excellent answer Pierre.

[Andrea_Bio]

Hi. Thanks for your help. Could I email you about your script as I think it will be useful for me but I don't fully understand it?

[Pierre Lindenbaum]

---

[Andrea_Bio]

thats very kind of you. I'll be in touch.

[Andrea_Bio]

Regarding this question, I'm not sure we're both on the same track which is probably my fault due to the wording in my question. I'm not asking whether exons can have different reading frames as you've already answered that question for me :) This is a simpler question. I'm asking, if an exon occurs in more than one transcript and has the same reading frame in those transcripts, is it possible that sometimes the exon is translated and sometimes it is not translated?