Within human genetics,

My current understanding of a transcript for a particular gene is that it's the exact nucleotide sequence and position of a particular instance of said gene in some individual.

My current understanding of an allele is that it's a less granular breakdown of the instances of a gene in a population. That is to say, one allele for a particular gene likely covers many (millions of?) unique transcripts for that particular gene, and groups them together due to their tendency to all have a similar phenotypic expression.

Is this correct?


I thought a transcript was the RNA sequence produced from transcription. The RNA copy of an allele that gets spliced and translocated to the cytoplasm for translation. Not sure if Wikipedia references are allowed here, but they seem to agree with my understanding.

An allele is a specific copy of a gene, i.e. you have two copies (alleles) of gene X; allele X1 comes from your mum, allele X2 from your dad.

I haven't heard of the less granular breakdown description of genes, so that might just be a different way of looking at it.

  • $\begingroup$ Yes I've seen the word 'transcript' used in that context as well, but I've also seen it used in the context of a particular 'gene transcript'. As for alleles, just reading from here: en.wikipedia.org/wiki/Allele#Multiple_alleles, it says "For example, at the gene locus for the ABO blood type carbohydrate antigens in humans,[6] classical genetics recognizes three alleles, IA, IB, and i, that determine compatibility of blood transfusions" But that can't be all, since basically every time you sequence another person's genome you should get a new allele. $\endgroup$ – Thoth Mar 12 '17 at 22:53
  • 1
    $\begingroup$ Not quite, an allele can be the same between different people and a person can carry two of the same allele as two copies of the same gene. $\endgroup$ – Oliver Houston Mar 12 '17 at 23:19

Alleles in the traditional sense represent genetic variation in a population at a given locus. Usually this is related to a given phenotype or observable trait. They can refer to areas of the genome that may or may NOT be transcribed into RNA, whether or NOT that RNA gets translated into protein.

Alleles can refer to a single nucleotide difference (SNP), or an insertion/deletion of one more more bases (indels). Thus for a given SNP associated with heart disease, if 97% of the population has a T and the other 3 percent have an A at that locus, then the alleles are the A and T alleles for that SNP.

Note: the variation could be in a regulatory region of DNA and never get transcribed. But the phenotype associated would be varied expression of a transcribed region it regulates.

A transcript is merely an antisense strand of a region of DNA with U replacing T. Note there are many kinds of RNA transcripts, not all of which get translated to protein. The transcript always mirrors the DNA below. If the DNA has genetic variation, the transcript will reflect it.

In the last 10 years, with the discovery of epigenetics, we also have started thinking about a genetic locus where the variation is a modification of base or histone that the base binds to as DNA is compacted. So some now also include methylation of a CpG as an allele, where one person might have 40% of CpG at that location methylated, and another have 67%. This difference can explain a trait difference and thus also be referred to as an allele (high methylation vs. lower).

Last but not least, much genetic variation results in no discernible phenotypes. Sometimes both Alleles still result in the same base being encoded during translation for example.



A gene is a unit of the genome stored in the form of DNA. It can be transcribed into mRNA (the transcript) to be used as a template in protein synthesis.

A gene can come in many different versions unique to an individual. That is, individuals can have different alleles, or versions of the same gene.

For example, in mice you can have many different fur colors. The color of their hair is determined by a series of biochemical reactions that are catalyzed by proteins. If each protein is functional, the mice will have a certain color of hair. If any of these proteins are not functional (because they are carrying a certain allele), then the mouse will end up being a different color than it would be if each protein functioned.

You can also have genes that produce somewhat faulty protein-products. As is the case with insulin-resistant diabetes.

  • $\begingroup$ But how come particular genes have so few recognized alleles then? Isn't there way more variation in a particular gene across the population than the 2-3 alleles commonly associated to it? $\endgroup$ – Thoth Mar 12 '17 at 22:57
  • $\begingroup$ Not really. Think of it this way. You are trying to assemble a car; you have ten sets of intructions, only one of the instructions will build you a functional car. How many types of cars can you build from these ten sets? $\endgroup$ – Bob Mar 12 '17 at 22:59
  • $\begingroup$ Two. It will either function, or it won't. $\endgroup$ – Bob Mar 12 '17 at 23:01
  • $\begingroup$ So you're saying that in a gene of say a million bp, that everyone will have one of ten possible million bp nucleotide sequences, without even a single SNP somewhere in there which deviates from one of these ten exact sequences? $\endgroup$ – Thoth Mar 12 '17 at 23:01
  • 1
    $\begingroup$ Not necessarily. Not all mutations have an affect on the protein-product. Because of things like wobble-base pairing, and the overlap in amino acids to trinucleotide sequences. And some mutations might not have that much of an affect on the product. Usually, for a phenotypic change to occur, a protein needs to stop functioning, or have its function drastically reduced. $\endgroup$ – Bob Mar 12 '17 at 23:07

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.