Morten Lauritsen wrote:
>> Hi all!
>> As usual, I have no idea if this is appropriate. If it isn't, please direct me.
>> I need programs that can do suboptimal alignments of sequences. That is,
> instead of just taking a shot at what's best, give me a graphic plot of the
> 5-10% best fits. This will give me an idea where the sequence aligns easily,
> and where it goes fuzzy.
This is not exactly what you ask for, but DotPlot in the GCG package
gives you a graphical output of local similarities between two sequences.
See http://biomaster.uio.no/GCGdoc/Doc.html for GCG documentation.
There are several reported methods for doing (at least almost) what
you ask for, but I dont know about available software.
> Also, I would like to see a program that can do local comparison of sequences,
> i.e. find small segments that match well, preferably many different.
There are several programs made for this purpose. Some do "local multiple
alignment", normally finding ungapped blocks. For example the Gibbs sampler
which can be downloaded by anonymous ftp from ncbi.nlm.nih.gov.
Reference:
"Detecting Subtle Sequence Signals: A Gibbs Sampling Strategy for Multiple Alignment"
Lawrence C.E., Altschul S.F., Boguski M.S., Liu J.S., Neuwald A.F., Wootton J.C.
Science 262:208-214(1993).
Alternatively you may want to look for patterns conserved between the
sequences. The program Pratt can find patterns of the PROSITE type
conserved in sets of unaligned protein sequences. Pratt can be downloaded
from ftp://ftp.ii.uib.no/pub/bio/
For more information see http://www.ii.uib.no/~inge/Pratt.html
Reference:
"Finding flexible patterns in unaligned protein sequences."
Inge Jonassen, John F. Collins, Desmond Higgins.
Protein Science 1995;4(8):1587-1595
Hope this helps. Cheers,
Inge.
--
Inge Jonassen,
Dept. of Informatics, University of Bergen, HIB, N5020 BERGEN, Norway
WWW:http://www.ii.uib.no/~inge/, email:inge at ii.uib.no,
tel +47 5554 4713, fax +47 5554 4091
