SNPs located in promoter region. OR exon, intron, 5'UTR, 3'UTR

Question

[wanghaho9797]

What happen if SNPs locate in promoter site?

and what about other regions? like exon, intron, 5'UTR, 3'UTR.

please explain how to change RNA and Protein if SNP locate in different site.

Answer 1

[dariober]

The questions is not very clear to me, what do you mean by What happen if SNPs locate in promoter/exon/intron etc? Do you want to predict the effect that a SNP will have on the protein or on the gene activity?

To annotate SNPs to genes and genomic features (introns, exons, etc.) and to predict the effect they will have in coding sequences there are nice tools, annovar is one of them. If the SNP is not in a coding sequence then predicting the effect is nearly (completely?) impossible. Promoters are not even well defined.

[wanghaho9797]

sorry my question was...

how would the variation affect RNA and protein if SNP is at the promoter? and also at the exon, 5'UTR, 3'UTR?

Answer 2

[Alex Reynolds]

You could access UCSC datasets and do genomic set operations is via the command line, using mysql, wget and BEDOPS tools like gtf2bed, bedops and bedmap.

For example, first get all human SNPs into a BED file:

$ mysql --user=genome --host=genome-mysql.cse.ucsc.edu -A -N -e 'SELECT chrom, chromStart, chromEnd, name FROM snp142Common;' hg19 > snp142Common.bed

Get genomic annotations and convert to BED:

$ wget -O - ftp://ftp.sanger.ac.uk/pub/gencode/Gencode_human/release_19/gencode.v19.annotation.gtf.gz \
    | gunzip -c - \
    | gtf2bed - \
    > gencode.v19.bed

You can take a subset of annotations in this file with awk (for example, exons) by reading the feature annotation keyword from the eighth column of the BED file:

$ awk '$8=="exon"' gencode.v19.bed > gencode.v19.exons.bed

Following this example subset, map annotations to SNPs, using bedmap to look for SNP rs IDs, whose SNP positions overlap those of Gencode exon annotations:

$ bedmap --echo --echo-map-id-uniq gencode.v19.exons.bed snp142Common.bed > exonsWithMappedSNPIds.bed

You can construct a BED file of proximal promoters by, for example, filtering for genes, finding the TSS by strand, and doing an asymmetric padding from the TSS start site.

For example, a promoter could be defined as 1 kb upstream of the TSS:

$ awk '$8=="gene"' gencode.v19.bed > gencode.v19.genes.bed
$ awk '$6=="+"' gencode.v19.genes.bed | awk '{print $1"\t"$2"\t"($2+1)"\t"$4"\t"$5"\t"$6}' - > gencode.v19.genes.forwardStrand.TSS.bed
$ awk '$6=="-"' gencode.v19.genes.bed | awk '{print $1"\t"($3-1)"\t"$3"\t"$4"\t"$5"\t"$6}' - > gencode.v19.genes.reverseStrand.TSS.bed
$ bedops --everything --range -1000:0 gencode.v19.genes.forwardStrand.TSS.bed > gencode.v19.genes.forwardStrand.TSS.1kb_promoters.bed
$ bedops --everything --range 0:1000 gencode.v19.genes.reverseStrand.TSS.bed > gencode.v19.genes.reverseStrand.TSS.1kb_promoters.bed
$ bedops --everything gencode.v19.genes.*.TSS.1kb_promoters.bed > gencode.v19.genes.1kb_promoters.bed

Then map to SNPs, as usual, with bedmap:

$ bedmap --echo --echo-map-id-uniq gencode.v19.genes.1kb_promoters.bed snp142Common.bed > 1kbPromotersWithMappedSNPIds.bed

Repeat for other annotation subclasses, as needed, or use any other annotation source (RefSeq, etc.) and repeat the same kinds of subsetting and mapping steps.

Answer 3

[H.Hasani]

As dariober already pointed out, there are tools to annotate and predict the effect of the SNP. In addition to annovar, there is SNPEff and VEP

A nice illustration for the SNPs effect and the meaning was given by VEP.

These tools give you as well, the position of the SNP in cDNA and final protein product so that you can give it to other tool; such as mutation assessor, mutation mapper and many other tools.

[wanghaho9797]

Thanks Hasani. I cannot use the tool.

Can you explain what affect amount RNA and Protein if SNPs occur in exon, intron, promoter?

I want to know generally.

[H.Hasani]

Basically and to start with, base substitutions are classified into synonymous and non-synonymous. If the substitution happens in coding regions it could lead to premature stop, or elongating the transcript (non-syn). The consequences could be sever that causes the disease or rises the risk to develop a disease.

More challenging is addressing the mutations in the non-coding parts of the genome, since its functionality not clearly understood yet. However, it is known that the effect of some mutations in promoter regions, or transcription factor could influence the transcription process or even its ablation.

Have a look at the GWAS catalogue, in which the trait associated SNPs are listed with the pubmed IDs. Also Wikipedia has nice bunch of articles about SNPs to help you introduce yourself to the topic.

Mutations with influence on the regulation process are called eQTLs (we recently published a study about them in the non-coding region). A nice study shows the distribution of trait associated SNPs in the genome is http://www.ncbi.nlm.nih.gov/pubmed/?term=19474294

Please note that my answer is REALLY an appetizer! Finally, you can use galaxy

Hth

[wanghaho9797]

Thanks for the answer.

Answer 4

[nbzhang]

Gene's expression could be regulated in multiple ways, eg. transcriptional and post-transcriptional regulation, translational and post-translational regulation.

SNPs in coding regions might result in variations at protein level;

SNPs in 5' UTR may alter start/stop codon of uORF therefore affecting the translation efficiency;

SNPs in promoter region that overlapping transcription factor binding sites would alter TF's binding affinity thus influencing gene's expression at RNA level.