Is this a snp?

[arkilis]

See the following image.

The first row is ref, the following two are reads.

Software in use: NOVO SNP



thanks~

[dpryan]

Probably. Our rule of thumb with Sanger sequencing was to require 2 reads in each direction. Presuming you use different amplifications, this will drastically decrease false positives.

[arkilis]

Probably. Our rule of thumb with Sanger sequencing was to require 2 reads in each direction. Presuming you use different amplifications, this will drastically decrease false positives.

Thanks for your reply. This is from only one amplification. You mean that if use another amplication in which has a similar result will lead to a snp? I thought this is just an insertion variant.

[GenoMax]

If there is a true SNP it should show up each time. Otherwise you may have a one-time sequencing error (it is looking too good to be an error, unless the scales in the chromatograms are not comparable).

[arkilis]

If there is a true SNP it should show up each time. Otherwise you may have a one-time sequencing error (that is looking much like a true SNP in the chromatograms).

well, it must have at least two reads to support it. If it is better to have another amplification to see whether can it happen again.

[swbarnes2]

I've looked at lots of trace files, and I'd call that a real, mixed SNP. So those are heterozygous samples, or you've got a non-clonal population of whatever that is.

Ideally, you'd have traces running forward and backwards across that site, and both would show it, but since your reference show no trace of that T peak, it's probably not an artifact.

[arkilis]

I've looked at lots of trace files, and I'd call that a real, mixed SNP. So those are heterozygous samples, or you've got a non-clonal population of whatever that is.

Ideally, you'd have traces running forward and backwards across that site, and both would show it, but since your reference show no trace of that T peak, it's probably not an artifact.

1. I think we define a SNP as a replacement not deletion/insertion. So this is not a SNP;
2. The quality can't strongly support it is a SNP. (At least two reads with good quality >Q30 to prove it)

Maybe I am not right, what do you think?

Or you probably can show a snp here. Thanks

[id0]

1. I think we define a SNP as a replacement not deletion/insertion. So this is not a SNP;
2. The quality can't strongly support it is a SNP. (At least two reads with good quality >Q30 to prove it)

Maybe I am not right, what do you think?

Or you probably can show a snp here. Thanks

You think that is an indel? I don't see how that could be an indel.

I agree about the quality. Half of the reads do not have it, so it's probably not real. You could always sequence again.

[dpryan]

1. I think we define a SNP as a replacement not deletion/insertion. So this is not a SNP;
2. The quality can't strongly support it is a SNP. (At least two reads with good quality >Q30 to prove it)

Maybe I am not right, what do you think?

Or you probably can show a snp here. Thanks

As id0 said, that's definitely not an indel. I see you've now updated the picture, from which I assume from the labeling that those are multiple samples. The quality of some of those (namely 131) is questionable, but the others are pretty clear-cut. I wouldn't focus too much on the exact quality in the context of Sanger sequencing. My general rule was to gauge things by whether they were visually obvious or not.

[swbarnes2]

Why did you change the picture? Now my original reply makes no sense.

You can not conclude a SNP is in the second one from the top, or the fourth one from the top. Those little peaks look way too much like background.

The third and fifth ones are mixed SNPs. Heterozygous, if you are looking at a single diploid organisms, or a mix of about 50/50 if you are looking at a population. There are no indels present, but for all I know, the poster who mentioned them was looking at yet another image that really had one.

[id0]

You can not conclude a SNP is in the second one from the top, or the fourth one from the top. Those little peaks look way too much like background.

The third and fifth ones are mixed SNPs. Heterozygous, if you are looking at a single diploid organisms, or a mix of about 50/50 if you are looking at a population. There are no indels present, but for all I know, the poster who mentioned them was looking at yet another image that really had one.

Agreed.

Sorry. I did not realize the image was changed. I was really curious how that would be called an indel.

[dpryan]

Sorry. I did not realize the image was changed. I was really curious how that would be called an indel.

The original couldn't have either

[arkilis]

As id0 said, that's definitely not an indel. I see you've now updated the picture, from which I assume from the labeling that those are multiple samples. The quality of some of those (namely 131) is questionable, but the others are pretty clear-cut. I wouldn't focus too much on the exact quality in the context of Sanger sequencing. My general rule was to gauge things by whether they were visually obvious or not.

Sorry for the update on the image, cause I thought this one is more representative. Yes, they are from multi samples.

My fellow and I have a debate on this position. He said it is very clearly a SNP which I do not agree. Also some reads here are have bad quality. To some extent, I only treat it as a heterozygous indel. What do you think? thx

[arkilis]

Why did you change the picture? Now my original reply makes no sense.

You can not conclude a SNP is in the second one from the top, or the fourth one from the top. Those little peaks look way too much like background.

The third and fifth ones are mixed SNPs. Heterozygous, if you are looking at a single diploid organisms, or a mix of about 50/50 if you are looking at a population. There are no indels present, but for all I know, the poster who mentioned them was looking at yet another image that really had one.

Sorry for the change on the image. Why are you so sure about it is a 'Heterozygous snp'?

[swbarnes2]

Because that's exactly what you'd expect a heterozygous SNP to look like. In the third and fifth traces, the otehr peaks look perfect, there is very low background in the other peaks, the alternate letter traces are the same height and shape as the wild-type letter, and you have an example of a clean, homozygous wild-type trace from these some primers, so it's not an artifact of these primers (as you might get if, say, this region was duplicated, and one region had the C, and the other the T).

The only thing which would make the case more solid would be having the trace from the other direction (or confirmation from another technology, obviously). But if you just don't have that, it's sound to conclude that the third and fifth sequences are heterozygous. I would not conclude that for the second and fourth traces. I would judge them to be homozygous C. That G peak thing is popping up all over the place in the fourth sequence, so that's an artifact, and the hints of a T peak are too small to be distinguished from noise.

I've seen tons of Sanger sequences that look like that, and occasionally, such traces have been accompanied by genotypes from other instruments (Illumina, Sequenom) which conform the heterozygous call. #3 and 5# are solid heterozygous samples. If they are the same sample from both directions, there is no question at all.

Edit: a heterozygous indel? No that's not at all what that would look like. With a heterozygous indel (and I've seen those) the forward read would be clean right up to the discrepancy, and then it would be peaks on top of peaks. And the reverse would look the same. And when you aligned them together with a reference, the forward would be clean on the left, and peaks on peaks on the right, and the reverse would be clean on the right, and peaks on peaks on the left.