Re: fastlink_4.1P-fix88-1_i386.changes is NEW (fwd)

[Elizabeth Barham <lizzy@soggytrousers.net>]

Andreas writes:

> Perhaps you could merge the contents into the style I used for all
> these pages?

Done. I'm satisfied with this one:

http://www.soggytrousers.net/repository/unknown.1

although I did not go into the datafile format is previously mentioned
because, after viewing a sample:

http://hpcio.cit.nih.gov/lserver/DOCS/FASTLINK/

(ch18test1.ped == "pedfile.dat" and ch18r395.par == "datafile.dat")
I've concluded that the manpage is not the correct place for
describing such a lengthy line of data.

   One thing I am hesitant to mention which seems to be the lack of
programming experience of the original authors. For example, when
feeding the data from the above cited web directory through unknown
and then ilink I received the following message:

--BEGIN--
YOU ARE USING LINKAGE (V5.1 (1-Feb-1991)) WITH  8-POINT
YOU ARE USING FASTLINK (V4.1P (6-Oct-1997)) AUTOSOMAL DATA

You are attempting to estimate allele frequencies.
You must have ALLELE_SPEED set to  0 in both 
unknown.c and commondefs.h, so that your alleles
are not mutated, then recompile UNKNOWN and ILINK.
FASTLINK will exit politely to allow you to change ALLELE_SPEED
--END--

which means, among other things, that they hardcoded some of the
variables which, I guess, could have been passed in on the command
line.

   Is dealing with this sort of thing our job?

   Elizabeth