1
$\begingroup$

I have predicted some SSR repeat on the gene of interest using SSRLocator program, which the result creates a question for me. Please consider the below sequence, which is part of gene sequence of interest (coding sequence):

GGTGATGAGGTAGAGGAAGAGGCTGAGGAACCCTACGAAGAAGCCACAGA GAGAACCACCAGCATTGCCACCACCACCACCACCACCACAGAGTCTGTGG AAGAGGTGGTTCGAGAGGTGTGCTCTGAACAAGCCGAGACGGGGCCGTGC

The program reported CCA motif as a repeat; since this repeat located on CDS, I expected to find the related amino acid track (prolin), but I observed the Threonine track in the corresponding protein sequence.Threonine is coded by ACC. Now I would like to know which motif is really a repeat, CCA or ACC. Is it possible to consider CCA as a repeat, but just on the transcript level not protein sequence or the real repeat is ACC and the software didn't work well here? Please share me any your opinion in this regard.

Thanks so much for your help and participation.

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

SSR searching tools have generally no idea about CDSs: they consider DNA/RNA sequences as strings and search for tandemly repeated patterns, nothing else. It is also important to remember that in many real-life cases different programs with different settings will reports differing results.

[Knowing that you always have two strands in the genomic DNA, one has to bear in mind that also SSRs can be "read" in either direction, such that for a stretch ACCACCACC the unit can be reported as ACC or GGT. Some programs by default will report SSR units sorted according to alphabet: ACC in this case. Now with coding sequences the story is a bit different: RNA has only one strand, i.e. it has a direction, which means that you cannot "flip" RNA sequences.]

mRNA besides having a direction, codes for protein with the protein sequences sometimes showing their own repeated patterns. SSRs in mRNA are mostly tri-nucleotides or with unit lengths divisible by three and coincide with the tandem aminoacid repeats. So, from the functional point it is a ACC|ACC|ACC|ACC|ACC|ACC|ACA (note that the last element is ACA) repeat, which codes for the respective T|T|T|T|T|T|T aminoacid repeat. To formalize this idea you search for SSRs in the RNA sequence and in the protein sequence and try to find the overlap.

From the point of view of the DNA-polymerase [the enzyme which changes the SSR length] it is a perfect CCACCACCACCACCACCACCA repeat: within these limits the fragment is prone to slippage. But for an actual slippage event the unit boundary does not matter. Imagine that the polymerase "skipped" a CAC unit in the middle: CCACCAC[CAC]CACCACCACCA. The outcome for the protein sequence will be the same: minus one threonine.

Improve this answer
$\endgroup$
5
  • $\begingroup$ Thanks a lot for your nice response. As DNA polymerase is often prone to slippage on the repeat sequences, so the repeat unit can be CCA unit as the software predicated. But, how about the situations that the RNA sequence and related protein sequences are matched? the CAG repeat unit is one of famous examples that generated poly-glutamine tract in the protein sequence that resulted in some human disease like Hungtingtun. Could you please help me to figure out what happened, how polymerase slippage occurred here? $\endgroup$
    – Mary
    Sep 20, 2015 at 10:25
  • $\begingroup$ If you have a repeated region, re-annealing of the newly synthesized second strand can cause priming of the DNA-polymerase at a different location. Somewhere in the middle it is no longer possible to "tell" whether the first unit started with C, A or G. This logic is implemented for example in Phobos: it can report "alphabetical normal form given by the alphabetically minimal string among all cyclic permutations of the unit", i.e. for SSRs with the first repeat units being AGC, GCA or CAG it will report AGC. $\endgroup$
    – alephreish
    Sep 20, 2015 at 10:42
  • $\begingroup$ Yes, of course. However, I'm a bit concerned about how I should reported the repeat unit on this sequence?! $\endgroup$
    – Mary
    Sep 20, 2015 at 12:58
  • $\begingroup$ As is: "such and such tool identified in this sequence an SSR with the following tri-nucleotide unit: ***, which corresponds to a stretch of seven threonine aminoacids in the translated sequence." The exact formulation for *** depends on the program: SSRLocator reports "CCA", Phobos reports "ACC". In both cases you can mention that the RNA SSR is shifted in relation to the protein repeat (if it is relevant). $\endgroup$
    – alephreish
    Sep 20, 2015 at 14:47
  • $\begingroup$ By the way, Phobos identifies here (correctly) an SSR with a length of 22: CCACCACCACCACCACCACCAC. $\endgroup$
    – alephreish
    Sep 20, 2015 at 14:49

You must log in to answer this question.

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