If PAM-1 matrix refers to time or evolutionary event for 1% change in amino acid sequence of a protein then how come there is PAM250? shouldn't PAM 100 be the highest where the two concerned sequences are 100% different

  • $\begingroup$ What attempt have you made to find the answer to this frequently asked question on line? Please read the help on asking good questions. Tell us what pages you have consulted and why you do not understand the explanation. Then we can help. $\endgroup$
    – David
    Mar 22, 2019 at 21:42
  • $\begingroup$ I used Xiong's and Jhumur Ghosh's bioinformatics books. they both explain how the matrix are made pretty well but my confusion is what I stated in my question. Page 148 Jhumur Ghosh's book states that the PAM 100 matrix corresponds to 100 accepted mutations per 100 residue. So, shouldn't it be the highest available matrix where the entire sequence is different. it clearly states there is 100% change in 100 amino acids after all. but then it also states one residues can also change more than once. how come one single residue can change multiple times when I am comparing two sequences only. $\endgroup$
    – Anindya
    Mar 22, 2019 at 21:53
  • $\begingroup$ English is not my first language, so, may be a little different set of words might be helpful for me to understand what really is going on. that's all $\endgroup$
    – Anindya
    Mar 22, 2019 at 21:54
  • $\begingroup$ No. The terminology is difficult even for English native speakers. The key is what an “accepted” mutation is. Try Wikipedia or other sources that explain that. I’m on vacation so can’t actually check. Search for “accepted point mutation” perhaps. $\endgroup$
    – David
    Mar 22, 2019 at 21:57
  • $\begingroup$ Accepted mutations are those that are accepted by nature, for example a hydrophobic amino acid substituted by another similar hydrophobic aa which does not negatively hamper with the protein structure and thus naturally selected. Any lethal mutations or silent mutations won't be part of accepted mutation. $\endgroup$
    – Anindya
    Mar 22, 2019 at 21:59

1 Answer 1


You misunderstood slightly what PAM1 means; it's actually (from Wikipedia) "1 [PAM] mutation to occur per 100 amino acids". So PAMn makes sense for any n, as in the 100 amino acids may mutate more than once each, on average.

Or if you put it in terms of PAM distance...

Is possible for two sequences to be more than 100 PAM apart? Yes: One position can mutate multiple times.

An attempted analogy: what does 600% casualty rate (for soldiers) mean? Well it means not only the original guy died, but his 5 replacements got (successively) killed as well.

  • $\begingroup$ If I am dealing with multiple sequences it makes sense if one position is getting mutated time and again but if I am comparing only two then how come one position is mutating multiple times? $\endgroup$
    – Anindya
    Mar 23, 2019 at 6:59
  • $\begingroup$ For example, if the query is MFPIMW and the sequence I am trying to align is MWPIMW then one spot at F is mutated, how can this specific spot mutate twice when I am dealing with only these two sequences. $\endgroup$
    – Anindya
    Mar 23, 2019 at 7:01
  • $\begingroup$ @Anindya: it can because a PAM is not a simple mutation. Given two randomly chosen amino acids, you may not have a PAM of 1 between them. Look at PAM1 for instance; note that the matrix entries are probabilities multiplied by 10,000 in there, i.e. the actual numbers are less than 1. $\endgroup$ Mar 23, 2019 at 7:05

You must log in to answer this question.

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