Nowadays it is possible to sequence the DNA of extinct species, such as the Neanderthals, the Denisovans, and others.

Is it possible to determine, solely from the sequenced DNA or from known bone fragments, how many chromosomes the individual (or the species represented by the individual) had?


This depends completely on the quality of the DNA. Since each chromosome is essentially a very long strand of DNA, breakages and missing sections are very common in extinct species due to degradation over time.

If a full DNA read is absent, no determination of chromosomal number can be performed.

Assuming a full read (covering all breakages) of the DNA is present, the telomeres of each chromosome can be found, and since chromosomes usually only have one set of telomeres each, this may be useful for chromosomal number determination.

Caveat 1: Chromosome 2 (human) is an example of how this method may fail, since a recently fused chromosome will still possess the tandem repeats characteristic of telomeres.

Caveat 2: Telomeres may vary in length due to the age of the cell, as well as be broken in multiple regions. Due to the tandem repeat nature of telomeres, it is nearly impossible to tell these different states apart experimentally.

  • $\begingroup$ If you had full coverage (I'm not sure how common this is), wouldn't you then be able to assemble across relict telomeres, such as in Chr 2? Presumably this was how they worked out the human Chr 2 was one chromosome in the first place? Similarly, I suppose another option (assuming full coverage) would be just to assemble and count how many sequences are produced (over size x). $\endgroup$
    – Sparhawk
    Dec 13 '14 at 23:48
  • $\begingroup$ Currently, detection of chromosomes seems to still be largely based on karyotyping and FISH. Relic DNA is usually highly fragmented on its own despite having full reads (100% coverage), and if the DNA happens to break between two telomeres in all the shotgun sequences taken, it would be nearly impossible to tell one from the other. $\endgroup$
    – March Ho
    Dec 13 '14 at 23:51
  • 1
    $\begingroup$ Yes, that makes sense. However, wouldn't this imply that the fragmentation would also prevent full detection of telomeres from sequencing? i.e. there might be fragmentation within telomeres, or joining telomeres to non-telomere regions? $\endgroup$
    – Sparhawk
    Dec 13 '14 at 23:57
  • $\begingroup$ Yes, it would. This is a nontrivial problem in sequencing, especially for tandem repeat regions, and the fact that telomere lengths change with the number of cell divisions only compounds the issue further. In fact, your point makes a good addition to the answer. $\endgroup$
    – March Ho
    Dec 14 '14 at 0:02
  • $\begingroup$ Thanks. One other thing I thought of would be to use synteny with close relations, but I guess that's cheating. $\endgroup$
    – Sparhawk
    Dec 14 '14 at 1:23

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.