non-intersecting alignments

[Gribskov, Michael]

Steve Modena apparently had some difficulty in understanding 

>In that last part there, what did he say?  I've noticed that a certain
>biostatistician that I occasionally work with always talks in similar
>cryptobabble with me.

an earlier posting (whose name I don't recall)

>>    The local alignment algorithm can locate matching segments within two 
>>sequences. The non-intersecting alignment means the matchings in the 
>>previous i local alignments will not be chosen as matching in the i+1 and 
>>later local alignments if we want to find K local alignments.
>>

For anyone who has followed the sequence alignment literature this 
should be fairly clear.  A non-intersecting alignment is one that shares
no pair of aligned characters (i.e. a matching) with the other 
alignment.  In other words, alignment i shares no matching with the i+1
or later local alignments.  Like most scientific specialties, the 
sequence alignment field uses a technical vocabulary that it greatly 
helps to learn.  This does not make it cryptobabble.  Abuse rarely leads 
to understanding (my opinion).

Michael Gribskov
gribskov at ncifcrf.gov



-- 
Domain: curtiss at umiacs.umd.edu		     Phillip Curtiss
  UUCP:	uunet!mimsy!curtiss		UMIACS - Univ. of Maryland
 Phone:	+1-301-405-6710			  College Park, Md 20742