Date:         Wed, 17 Nov 1999 10:05:29 -0800
Reply-To:     Stephen Arthur <sarthur67@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Stephen Arthur <sarthur67@YAHOO.COM>
Subject:      Re: Stepping through a sequence of data until a criteria is met
Comments: To: peter.crawford@db.com
Content-Type: text/plain; charset=us-ascii

This is my reply to Peter Crawford that I am sharing
with the list.

--- peter.crawford@db.com wrote:
>
>
> There are some really effective and simple things
> that you can do with character
> variables and functions, but...
>

That is what I was thinking (maybe deleting the first
observation after a test fails, and then continue with
the step from there), but...

> How do your items arise?
>    separate variables 1 char long
>    one long string
>    separate observations
>    external text file
>    macro variable

The data is imported from a gene sequence file a
scientist with a 'SEQ File' format.

This data was orginal one long sequence of 1276
characters (ATCG).  The scientist is interested in
certain segments along this sequence.

I separated the sequence using the replace function in
word to replace 'A' w/ ' A ' (I am doing this on the
fly) to create separate observations.

>
> What would the likely maximum be for that arbitrary
> 73 ?
>    no more than 200
>    frequently over 200, but under 32768

This will be the case:

"frequently over 200, but under 32768"

>
> What output do you want from the algorithm ?
>    one observation for each set of 50 bases
>    one observation per accepted sequence
>    the first accepted sequence
>    the position where the first accepted sequence
> starts
>    a macro variable holding the first accepted
> sequence

The output I would like is a sequence of 50 bases from
a segment containing 50% or more CG content.

From there I need to find a 45 base segment that is
contained in the 50 base segment, all the way down to
a 10 base segment from the previous 5 base step w/ at
least 50% CG content.

>
>
> Fascinating question -- what is the application
> generating this requirement ?

The 50% CG rule requirement meets a criteria for
picking a possible probe the scientists can use to
direct their experiments.

Thanks for your help.

I need to think this through more.

steve



__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com