1
$\begingroup$

My background is not genetics. 2.I am not interested in knowing how dna sequencing or genotyping is done. 3. I am interested just in the nature of the results as described here.

Now coming to the question: Humans have two sets of chromosome in each cell. So if I say sequencing chromosome1, what does that actually mean? Which of the two chromosome's gets sequenced? Of genotyping-do comparison take place between two+two=four chromosome1's? Or is my entire undertanding flawed?

$\endgroup$
  • 4
    $\begingroup$ What is your own research about this question? To avoid closure as "homework question" (which is also used as a synonym for low effort questions), please add this information using the edit function. $\endgroup$ – Chris Aug 16 '16 at 11:36
  • $\begingroup$ From your question I gather that you do not have a basic understanding of sequencing....The answer can get pretty long....It would be advantageous to know your background and as Chris mentioned to know your research on the matter. $\endgroup$ – FoldedChromatin Aug 16 '16 at 11:51
  • $\begingroup$ No I don't.My background is electronics engineering. But I am keenly interested in this topic and is eagerly waiting for the answer. Please do explain in simple words how it is done and more importantly about the two strands getting sequenced. $\endgroup$ – crusoe Aug 16 '16 at 11:59
  • $\begingroup$ genomenewsnetwork.org/resources/whats_a_genome/Chp2_2.shtml. I have a rough idea that dna is cut into pieces ,stained etc. What I am concerned about is not the process,but the nature of the output. In what order will the output be presented? Or is it completely random? $\endgroup$ – crusoe Aug 16 '16 at 12:21
1
$\begingroup$

I am interested just in the nature of the results as described here.

By aligning this sequence to all nucleotides of all organisms, we would realize that the given (short) sequence occurs in various organisms besides humans. Therefore it might come from a species with only one chromosome see NCBI Blast !

For the sake of the example, let's assume that the species had multiple instances of each chromosome (e.g. one by human mother, and one by human father), and look at "the results as described here". As no detail is given on the protocol, let's further assume the most common scenario - namely that there was no experimental distinction or separation of the individual instances of chromosomes.

Looking at the results, we would realize that is not "next generation sequencing", but Sanger-sequencing.

Now we look at the provided scheme again (note: this is not real experimental data): clearly the bands on the left side only occur for one letter/base, and at a given position of the sequence, there only is one very clear peak on the right hand side (as opposed to the possibility, to have multiple peaks corresponding to different bases).

We would have to conclude that every instance of the chromosome has an absolutely identical sequence. Thus we would conclude that it would not be important to distinguish individual instances.

But wait - aren't maternal and paternal chromosomes different? Absolutely, the given example however only looks at 25 bases rather than the full chromosome - and it would quite possible that they are absolutely identical. (note: there are various experimental techniques to select only specific regions of DNA or RNA for analysis).

$\endgroup$
1
$\begingroup$

I would say your understanding here isn't necessarily flawed - rather, there's a little trick being played on you.

The figure depicts typical results of Sanger sequencing. This type of sequencing always requires a so-called primer, a short sequence of around 20 bases which are known. The code following those 20 bases will be sequenced. Using this technique alone, you can't practically sequence a whole chromosome - it's usually only good for about 700-1200 bases after the primer.

The bands in the graphic on the left and the peaks on the right side correspond to the base at position "primer + n" (n being the number of bands/peaks starting to count at the bottom of the graphic).

Now consider that in practice, this technique is performed on a liquid sample containing of course a huge number of DNA molecules. In an ideal scenario, every single molecule is identical. In that case, the sequence behind the 20 primer bases is identical and you will get clear bands or peaks as shown in the graphic.

When you take a sample from an organism like a human, there will be a variety of DNA molecules in the mix.

What gets sequenced? Simple, everything that sits behind the primer you're using to sequence. If different molecules have different sequences there, the result will be bands in different columns at the same position on the gel (left side of the graphic) or two different-coloured peaks at the same location in the chromatography (right side of the graphic).

Example: Genotyping

We consider the XYZ gene. There are two variants of this gene in humans, and the only difference is at position 167 of the gene, where one variant has an A and the other a G. Thanks to the human genome project, I know the full sequence of the XYZ gene, so I design my primer to span positions 50-70 of XYZ. Therefore, my sequencing result should yield the XYZ gene, starting somewhere around position 100 (even the highest quality sequencing reactions miss out a few bases after the primer). If I take a sample from a human who has the same XYZ variant on both sister chromosomes, I will only get a single sequencing result - either with A or G at position ~67 of my sequencing. If my sample came from a human who has one variant on one chromosome and the other variant on the other, my sequencing will be ambiguous at position ~67, and upon closer inspection I sequenced two variants from this sample: one with an A and one with a G at this ambiguous position. This way I can determine a human's genotype for this gene: homozygous XYZA/A, XYZB/B or heterozygous XYZA/B.

PS: Don't let the double-stranded nature of DNA confuse your thinking here, Sanger sequencing always only sequences one strand (the one that you designed the primer for)! So two sister chromosomes = two double strands = four single strands, of which two will be sequenced because the other two contain the complementary sequences.

PPS: The way this works can cause huge problems if the primer isn't designed properly. With 20 bases, it's possible that the same 20 bases appear elsewhere in the genome, and suddenly the sequences following the primer aren't only different in one or two positions, but everywhere.

$\endgroup$
1
$\begingroup$

That's a picture of Sanger sequencing. If that sequencing had been done in a region where a diploid organism was heterozygous for a simple SNP, the trace file would instead of having single clean peaks, would at the point of the SNP have two half sized peaks of two colors for the two letters present at that site. If the difference between the two alleles was more involve than a single nucleotide polymorphism, the trace file might look quite different.

So if I say sequencing chromosome1, what does that actually mean?

People don't often just sequence a whole chromosome like that, and pretty much no one would do it with Sanger technology anymore. But if you were sequencing a diploid non-inbred organism, you would expect at points of heterozygosity to get a mix of signals.

Of genotyping-do comparison take place between two+two=four chromosome1's?

Diploid organisms have 2 copies of a given chromosome, not 4. Each chromosome has two strands, but the information on complementary strands is identical.

$\endgroup$
0
$\begingroup$

UPDATE


Before coming to sequencing you have to consider that you are not sequencing a single chromosome from a single cell. You are sequencing a DNA sample corresponding to a population of cells, containing all the chromosomes. I do not know of a procedure to target a single chromosome, targeted procedures for specific regions on the genome sorrounding genes are in existence.

Or is my entire undertanding flawed?

Yes, I would say lacking not flawed.

That said,

So if I say sequencing chromosome1, what does that actually mean?

Hypothetically that would mean you are sequencing DNA from chromosome 1 corresponding to a population of cells. Meaning you are retrieving the sequence of chromosome 1 for usage in downstream analysis by polymerizing its complementary DNA sequence by taking advantage of the procedure of DNA replication.

Which of the two chromosome's gets sequenced?

There is no way to differentiate between two chromosomes in mainstream sequencing experiments. The cell does not have an idea about which chromosome is which and neither do you, so both chromosomes get sequenced at the same time. Please note, I said no way to differentiate between two chromosomes in mainstream sequencing experiments

do comparison take place between two+two=four chromosome1's?

This has two parts. What do you mean by four? If you are implying that chromosome 1 has two strands and that equates to 4 then no, the two strands together make one chromosome. Your cell has 2n or two chromosome 1's. If you are sequencing a population of cells (let's say 1000), you have two thousand chromosome 1's which are being sequenced all of which are present in pairs across 1000 cells.

Do comparisons take place?

In mainstream experiments we do not compare. That said, methods for differentiating between the two chromosome 1's exist. The oldest I could find is an article by M Nagano on allele specific sequencing. In this case, since your parents equally contribute towards your DNA, each of them carry some mutations specific for their DNA. Using this feature we can differentiate between the paternal and maternal chromosome 1.

There's also single cell sequencing, in this case you would be sequencing the two chromosome 1's from a single cell, theoretically you can then also do allele specific sequencing on a single cell allowing you to differentiate between the two chromosome 1's present in the cell.

Finally, there is no way to differentiate the two strands, after sequencing when you align your data back to the reference genome, you will get to know the strandedness of the data.

Answer to old question


Nowadays the word sequencing is synonymous with High-throughput sequencing, and the article you linked to in the comment relates also to high-throughput sequencing techniques (I got this from a cursory glance).

While sequencing dna which of the two strands get sequenced?

Both get sequenced.

In what order is the results provided

Random

while sequencing how are these two strands sequenced?

If it is single end sequencing, we do not know which strand was sequenced first. (There are ways, but for simplicity's sake let's not go there)

If it is paired end sequencing, one read originates from the sense strand and the other read originates from the opposite strand.

You can watch a youtube video here to find out what is the difference

Is it that result of one strand is provide first followed by the other or is it that only one of the two strands get sequenced?

Both are reported.

The question you have asked is very broad, I can go on for a handful of A4 pages and still not finish. Coming to your first question, I said both get sequenced, because we have no way of knowing while sequencing which strand is getting sequenced (there are stranded sequencing techniques, but you should digest what is mentioned/linked here first and later on with some more research come back with a question on stranded sequencing).

After sequencing you get your result in random order, because what you get is not really result but more raw data files which are called FASTQ files. Read up on FASTQ a bit. After sequencing you don't really get a FASTQ but you get a BCL or base call file, which must be converted to FASTQ. These FASTQ must then be either filtered or not based on quality then aligned to a reference genome. This is where you get to know which read came from which strand.

You should read up more on single end sequencing and paired end sequencing to better understand how DNA is sequenced. Also check out this video. This is the most simplified version which I could find. It should make you curious without bombarding you with too much information.

Last of all, in a single end sequencing you have no way of knowing without alignment which strand got sequenced, but in paired end (where a DNA fragment is sequenced from both ends, generating two paired reads) generally, one mate aligns to the sense strand while the other aligns to the antisense strand.

$\endgroup$
  • $\begingroup$ I do not know of a procedure to target a single chromosome Check out single cell sequencing. I'm more familiar with it in terms of RNA-Seq, but one can perform single-cell genomic sequencing as well. $\endgroup$ – MattDMo Aug 16 '16 at 17:54
  • $\begingroup$ I meant that I do not know of any procedure till date, which isolates an entire chromosome for sequencing...there are targeted protocols for sure...but targeting an entire chromosome? I'm yet to encounter a procedure which employs isolation of a chromosome before sequencing. $\endgroup$ – FoldedChromatin Aug 16 '16 at 18:09
  • $\begingroup$ OK, I understand. I doubt such a thing is possible in somatic cells, as there is no way to differentiate between maternal and paternal chromosomes. However, I don't think there is a scientific need to do that. If you want to know where a particular SNP or other mutation comes from, just sequence the subject's parents (if possible). $\endgroup$ – MattDMo Aug 16 '16 at 18:16
  • $\begingroup$ This poster is asking about Sanger sequencing, not next gen. There are no bcl files in Sanger sequencing, and it's not clear to me that returning Sanger data in FASTQ format is standard. It wasn't in the years that I worked in Sanger sequencing. $\endgroup$ – swbarnes2 Dec 19 '16 at 19:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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