I would like to understand how a gene is assigned to a specific coordinate (e.g. NOC2L Chr1 879583(Start) 894679(End)). Next, let's say NOC2L gene has a few different isoforms, do those isoforms have different start and end site? What is the definition of start and end site of a gene?
$\begingroup$
$\endgroup$
1
-
1$\begingroup$ You are asking several questions here, and on this site we can prefer to deal with one at a time for indexing purposes. One of your questions is the definition of the start and end of a gene. This question is easily answered in terms of the transcript, but more difficult if one considers enhancer sequences. I suggest you read about this. Your other question is about isoforms, which arise in two ways, either by gene duplication and divergence or differential splicing. Again these topics are well covered in the literature. As for assigning co-ordinates, that is just mapping. $\endgroup$– DavidCommented May 19, 2021 at 11:50
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
4
The genes are defined by a reference genome - one that is considered the standard for comparison to. This reference is defined by which build of the genome you are comparing to, so there are different builds based off different assemblies of information. These are updated as new, more complete sequencing of the genomes (e.g. low complexity areas etc) are completed. At the moment for humans we are up to GRCh38 (Genome Reference Consortium, human build 38).
In general most gene positions won't change significantly between builds as these are fairly well defined - we can recognize promoter sequences, and the start condon (ATG) and stop codons (TAG, TGA or TAA) too. The sequence of the genes is largely conserved, so by comparing several sequences of the same area, we can see which bits are conserved within that sequence and thereby identify if it is a gene. We can also compare to sequenced cDNA (from mRNA) which is the transcribed form of the DNA and identify the sequence from that.
With respect to isoforms; it depends on the isoform, sometimes this form has a shorter or longer sequence, but often it is an altered (but same length) sequence that a proportion of the population carries. However, generally most of the sequence will be the same or very similar to the reference gene.
-
1$\begingroup$ With all due respect for your attempt I must point out the weaknesses in this answer as someone upvoted it. I don't see that reference genomes have anything to do with this. People were looking at genes for different isoforms long before reference genomes. Also start and stop codons for translation are not relevant to the ends of the gene. And some isoforms result from gene duplication and divergence and others from differential splicing. $\endgroup$– DavidCommented May 19, 2021 at 11:45
-
$\begingroup$ @david - The OP asked how the start and end positions were defined on the chromosome, I simply gave the most parsimonious answer based on the likelihood that the OP has a bit of a limited understanding of the topic. AFAIK this is how it is done these days; no longer assigned relative to banding in a karyotyping exercise. $\endgroup$– bob1Commented May 19, 2021 at 20:32
-
1$\begingroup$ This answer does not mention the word "co-ordinate". Nor does it explain how the co-ordinates were defined in the chromosome, given that the ends are not sequenced. The title of the question suggests that the main concern of the poster is isoforms and asks about starts and ends. This answer does not address this. $\endgroup$– DavidCommented May 19, 2021 at 21:38
-
$\begingroup$ Thanks a lot for all of these responses, I apologize for responding so late, and yes it's true that I have very limited understanding about this. I raise this question because, I have a data from NCBI that tells the position of each gene (like the example I've given in the post). However, I heard from my colleague that gene isoform can actually have different start and end position. Thus, I am wondering why NCBI just give us 1 fix position instead of giving us all the known isoform start and end position (if it's really different). $\endgroup$ Commented Jun 22, 2021 at 7:31