A quote from a review paper:

ALDH7A1 gene (discovered in 2006), which is located at chromosome 5q32.2, contains a transcript of 4964 base pairs and 539 amino acids divided among 18 exons [1,10,16,17].

As I understand, a gene cannot contain a transcript, it can only encode an RNA transcript. I wonder whether the authors meant that the transcript-encoding sequence of DNA constists of 4964 base pairs, or that the RNA transcript itself consists of 4964 base pairs.

I think it's the first, because I more often read about DNA as consisting of "base pairs", but I've just read Wikipedia and started having doubts, because RNA also seem to contain base pairs. I haven't been reading about this a long time, so I'm confused.

  • $\begingroup$ I suspect that you're getting wrapped up in the terminology. Think about amino acid coding- the 539 amino acids * 3 = 1617 coding bases only. The transcribed portion of the gene might have 4964 base pairs. People get a bit loose with terminology, which can be annoying, but you can always go back to the iron(ish) laws of molecular biology. $\endgroup$ Commented Dec 21, 2022 at 15:51
  • $\begingroup$ @MaximilianPress — Genes encode RNA. mRNA encodes protein. The poster is correct in my opinion in saying that a gene encodes one of more transcripts. The code is dA=rA, dG=rG, dC=rC, dT=rU. Additional coding determines splice sites. $\endgroup$
    – David
    Commented Dec 21, 2022 at 19:08
  • $\begingroup$ @David ... I guess. I suppose that I am responding more to genes "encoding" transcripts to the exclusion of other forms of relationship ("contain" is obviously chemically wrong, but from a genome annotation point of view is fine). Regardless, I'll delete my second comment as at best distracting from the real point, and at worst incorrect. $\endgroup$ Commented Dec 21, 2022 at 20:19
  • 1
    $\begingroup$ I think it better to say "a gene…specifies" and avoid "encodes". "Contains" is sloppy for a professional scientist, but it happens to the best of us. $\endgroup$
    – David
    Commented Dec 21, 2022 at 23:31

2 Answers 2


It's the second: The RNA transcript of the gene (before editing) has a length of 4964 bases. The use of "base pairs" is wrong in this context since the transcribed RNA has no paired bases. This is a feature of the DNA.

This transcript (or better: the processed exon parts of it) code for the amino acid sequence of the protein which is translated from the mRNA.

enter image description here

Image from here.

  • $\begingroup$ I think the diagram is misleading. It suggests, wrongly, that the whole of the transcript exons encode amino acids, and compounds this by having the transcript start with AUG. (And there is a typo: "transcript are code for") $\endgroup$
    – David
    Commented Dec 21, 2022 at 23:35
  • 3
    $\begingroup$ @David eh, the image is clearly not trying to be rigorously accurate. It's a doodle. I actually like it since it gets the main point across, let's not get hung up on the lack of UTRs and introns or the fact that no protein will ever be just three amino acids etc. $\endgroup$
    – terdon
    Commented Dec 22, 2022 at 9:53
  • $\begingroup$ @terdon — The diagram may be appropriate in an answer to a question with a low degree of sophistication, but hardly to one that includes an explicit mention of exons. $\endgroup$
    – David
    Commented Dec 22, 2022 at 14:46

Caveat Emptor

Scientific writing should be as precise and unambiguous as possible, and the poster is correct in implying that the language in the cited review is incorrect. A gene certainly does not contain a transcript. The sloppiness is further borne out by the incorrect reference to the length of the transcript in base pairs, as pointed out by @Chris. Finally, 4694nt is incorrect for the length of the transcript: it is 4765nt.

One can only say that this happens, especially in cases such as the present in which the authors appear to be specialists in brain function rather than molecular biology, are being obliged to write in a language other than their native one; and this is compounded by journals with inadequate editorial supervision.

Even worse, is the fact that mistake in citing the length of the transcript in base-pairs appears in the ENSEMBL entry for this gene, although not in that from NCBI, which gives the lengths in nucleotides (nt). I would, however, take issue with NCBI’s use of the term encoded in “The protein encoded by this gene…”, as a protein is encoded by a mRNA, not a gene, and the primary transcript of one gene can result in several different mRNAs, with different protein products — as in this case. I would recommend the use of the word specify rather encode in this case.

One must remember, however, that both ENSEMBL and NCBI employ temporary staff with qualification outside of molecular biology to compile these gene descriptions.

How to find gene details oneself

There are standard genomic resources where one can fairly easily find the details of any gene oneself, especially, as in this case, human genes. The two main sites are NCBI and ENSEMBL, for which the links to this gene are already posted above. To understand the data, one needs the basic knowledge of gene structure, transcription, splicing and translation of mRNA expected of those posting questions of this sort. From these sources I would list the salient information about ALDH7A1 as follows.

  • The human gene ALDH7A1, aldehyde dehydrogenase 7 family member A1, is located on chromosome 5.
  • It runs from nt 126,541,841 to 126,595,219 on the reverse strand, and hence it encompases 53,379 bp and its initial transcript are 53,379 nt.
  • The initial transcript contains 18 exons, which constitute a mature mRNA of 4765 nt (not 4964) after the introns are spliced out.
  • This mRNA is translated into (or “decoded to”, if you wish) a protein of 539 amino acids, the difference between the 1620 nt coding sequence (including the termination codon) and the 4765 nt of the mRNA being the untranslated 3′ and 5′ regions of the mRNA.

This is illustrated in the diagram of the gene below in which the vertical bars or boxes are the exons, with the translate regions solid and the untranslated regions blank.

ALDH7A1 gene structure

Is this really the gene?

It seems to me that calling this the ALDH7A1 gene is rather dubious — or misleading, at least. The gene that gives rise to the transcript described above also specifies 36 other transcripts, many of which appear to be translated to different variants of the alcohol dehydrogenase protein. (See the transcript map in the page cited above.)


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .