I'm looking for the lengths of the centromeres of human chromosomes.

The best I could come up with so far has been: The length of individual centromeric arrays was found to range from an average of 680 kilobases for the Y chromosome to 3000 kb for chromosome 11.

Here they say that "Human centromeres are located at repetitive alpha satellite DNA arrays that compose approximately 5% of the genome." Which unfortunately is not the same as saying the centromeres compose 5% of the genome. (Or maybe it is? Biologists to the rescue!)

Ideally an answer would contain the length of the centromere in each chromosome, but a precise percentage for the whole human genome would be good enough.


No specific length for centromeres. Actually in a real chromosome in a cell, centromeres are usually very condensed sequences of DNA, they are winded and folded (if you know about the DNA structure). Thus you cannot measure the length of the centromere unless you decondense the chromosome first.

  • $\begingroup$ My question is about the length in base pairs. My understanding is that you can sequence centromeres, you just cannot assemble them. But the sequencing reads might be enough to give an estimate of the total base count. $\endgroup$ – BlindKungFuMaster Apr 19 '16 at 7:35
  • $\begingroup$ @BlindKungFuMaster There's not much genes in centromere region. So it is hard to design a RNA primer for sequencing. Sequencing cannot be done on the entire DNA molecule due to the the vulnerability in the structure of macromolecules. By the way, why does total base count matters? What triggered this question? And what do you mean "cannot assemble them"? $\endgroup$ – Young Apr 19 '16 at 15:12
  • $\begingroup$ My question is in the context of whole genome assembly. I wanted to know how much of the human genome cannot be reconstructed, when treated as a DeNovo Assembly. The centromeric region is a big chunk of the unassembled regions in the human genome, I just wondered how big. $\endgroup$ – BlindKungFuMaster Apr 20 '16 at 8:01

There is no fixed length of human centromeres.

As the paper you cited says in its introduction, there are array-length polymorphisms in centromeres. A polymorphism can be described as two or more forms of the same DNA sequence existing; which one you have depends mostly on which one you inherited from your parents. There are also polymorphisms that are created by imprecise DNA duplication during meiosis. Array-length polymorphisms can be explained by the latter.

As the DNA in centromeres is highly repetetive, it can happen that during replication, the number of repeats is decreased or increased. This leads to many different copy numbers in different individuals. Therefore, your answer cannot be answered with a specific number.

  • $\begingroup$ I would settle for an average number, a range, or the specific number of a specific individual. $\endgroup$ – BlindKungFuMaster Mar 18 '16 at 12:02

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.