[ssml] Re: [BiO BB] Re: BiO_Bulletin_Board digest, Vol 1 #601 - 3 msgs

Dan Bolser dmb at mrc-dunn.cam.ac.uk
Fri Jan 9 04:26:45 EST 2004


++ sravan  sravan--
>
> Hello everybody,
>          I am working with secondary structure alignments. In
> order to optimize the gap penalties for the alignment of secondary structure
> strings with dynamic programming mehtod, I am using the following method.  
>     1. Selecting of N number of pairs of proteins(each pair is a set of
> two structurally similar / sequence similarity: 70<%id<100) ).    
>   2. secondary structure generation of the chains per pair.    
>   3. Running the alignments of each pair with N combinations of gap
> penalties.(structure specific).       4. At wchich combination
> various proteins give highest structure alignment(sst aligns) with the paired one
> than with the others(low scores), that combination can be selected. queries:::
> -----------
>  00) I am trying to select the similar pairs from FSSP table given by DALI.
> (TABLE2.html). In this pairs given in table I wanted to select the unique pairs
> based on a criteria (rmsd,%Id)..
>       CAN I SELECT A PAIR WITH SEQ %ID: 70<%ID<100? WHAT
> SHOULD BE THE VALUE OF RMSD?


I would get all sequences in a fold then compare sequence identity, then pick pairs
from within the fold which match your criteria.

>  (01) ANY IMPROVEMENTS TO THE METHOD I AM THINKING?

70% sequence identity is very high! -


>  (01)ANY OTHER METHODLOGY TO OPTIMISE THE GAP PENALTIES FOR SECONDARY STRUCTURE
> ALIGNMENT?


Probably many you could think of (i.e. counting the gaps between homologus pairs of
SS in structures), but none that I know of.

>
> Thank you all for the co operation.

Cheers!



> Regards.
> P.Sravana Kumar.
>






More information about the ssml-general mailing list