problem with reading in combined genotype call file

[Sunny]

Dear Dr. Li,

I'm using dChip (build date: 4/9/2009) to do CN/LOH analysis of Affy
500K data. I have a group of CEL files and a combined genotype file
for each chip type (Nsp and Sty). I used Analysis/Open group to read
in CEL files first, then read in genotype call file using Analysis/Get
external data following this instruction:

[V5/29/06+] To read SNP CEL files with combined genotype file but no
individual matching TXT/CHP genotype files, first prepare a combined
genotype call file from Affymetrix genotyping software, in this format
(save in text format in Excel). Make sure the column names are the CEL
file names without “.CEL” extension. Then after “Open group”, use
“Analysis/Get external data” and specify this genotype call file as
“Data file” and check “Read SNP call file and save to DCP file”. In
future sessions of dChip, “Open group” will be fine without doing “Get
external data” again, since genotype calls are already in DCP files.

The program run through successfully, and the dcp files and an array
summary file were created, but I doubt that the genotype calls have
not been integrated into the dcp files or have not been read in at
all, that based on my these findings:

1. In the array summary file, the SNP call % is 0 for all arrays.
2. When I do Analysis/Normalize, all the arrays shown 0 genotype call
rate.
3. The size of dcp file was not changed after reading the genotype
call file.

My questions are:

1. How to check to make sure if the genotype calls have been read
into the dcp files?
2. If the genotype calls were not read in, what could cause this
problem?
3. My normalized CEL value and computed MBEI are still valid even
without genotype call read in, is that correct?
4. If I have to read in individual CHP files for genotype calls, do I
have to redo everything, i.e., redo the CEL input, Normalization, and
computer MBEI? or just read in CHP files in an additional single step
and the program will automatically save to dcp files? And how to do
that?

Sorry for the lengthy message. I'm a new dChip user. I really
appreciate any of your advices.

Best regards,

Sunny

[Cheng Li]

Sunny,

Could you attach dChip output messages after you do "Analysis/Get
external data"?

3. Yes

4. CHP should be read together with CEL files, so CEL files have to be
renormalized, etc.

Cheng

[Sunny]

Cheng,

Thanks for your prompt reply.

I run Analysis/Get external data yesterday, and closed the session.
Is there any way to search the output messages in histry? I remember
that during the running, message show like: reading ...file, and saved
to dcp file, ..., there was no error message during the whole process
and the run was successfuly finished.

I thought that the CEL files for CN analysis, and the genotype call
for LOH analysis, can I just finish the CN analysis first with the
normalized data I already have, then imput the CHP files later on? or
the CHP files can only be read in with CEL files at the very beginning
step?

I really don't want to redo everything if I can figure out why reading
the combined genotype file should be a problem. Also to redo it, I
need to change all the CHP file names to match the CEL files. Do I
have other things to check to solve the problem?

Thanks again.

Sunny

> Sunny- Hide quoted text -
>
> - Show quoted text -

[Cheng Li]

-- Forwarded old discussions

Posted: 31 Jul 2007 09:20 pm Post subject: is it possible to use
external file for call data?

I have cel files (for intensity) and genotype call file. When I try to
merge genotype call files into DCP files using “Analysis/Get external

data” and specify this genotype call file as “Data file” and check

“Read SNP call file and save to DCP file” the program successful reads
in dcp files, reads in SNP call file, but crashes when it tries to
save the SNP call file to the dcp file.

I have followed instructions for 5/29/06 note, as above. I appreciate
any advice about how to integrate external call files and DCP files.
_________________
Thanks, Diana

Posted: 01 Aug 2007 02:52 pm Post subject:

What SNP array type are you using? Could you attach the screenshot of
"Get external data" dialog and the dchip output after "Open group" and
"Get external data" (the latter can be screenshot of crash).

Posted: 02 Aug 2007 10:21 am Post subject: screen shot

I am working with sty Affy 500 arrays.

I have attached screenshot. Also, when DChip tries to write the
external file to the .chp file, in addition to crashing the .dcp file
for the first subject is erased.

Thank you very much for your help.
_________________
Thanks, Diana

Posted: 05 Aug 2007 03:04 pm Post subject:

This is a bug of saving dcp file when probe data are not loaded in
memory. Please use this 8/5/07 version to correct it:
http://www.hsph.harvard.edu/~cli/dchip2006.exe

Thank you very much for helping to correct it.


Posted: 08 Aug 2007 08:29 am Post subject: read external call data
into dcp files

--------------------------------------------------------------------------------

The program now goes through the step of reading the external call
data into the dcp files, but I still get "O" for SNP call % and no A/B
calls show up when I run the data.

-------------------------------------------------------------------------------

The array summary file is probably generated before reading in SNP
calls. You can redo "Open group" to see if call % is updated. Or try
"Analysis/Chromosome" to view genotype calls.


Posted: 09 Aug 2007 04:50 pm Post subject: Bug corrected in
V6/29/07-8/8/07: reading SNP genotype file

Charlotte,

Thank you very much for reporting the bug and helping to locate it
between close dchip versions.

It's due to error in reading reference normal or combined genotype
file by dchip versions from 6/29/07 to 8/8/07. It thus affects these
functions:

-- LOH analysis (LD-HMM and haplotype correction)
-- Reading external genotype call file for CEL files without paired
genotype call file (e.g. SNP 5.0 and 6.0 data).

Please use this 8/9/07 version to correct it:
http://www.hsph.harvard.edu/~cli/dchip2006.exe

Thanks again,

Cheng


From: Charlotte Schjerling
Sent: Thursday, August 09, 2007 10:14 AM
To: Cheng Li
Cc: Charlotte Schjerling
Subject: LOH considering haplotye not functioning properly since June
28 2007

Hi Cheng

I have some problems with the latest versions of dChip as the LOH
considering haplotype function is calling many many small regions as
LOH in dChip versions later than June 28, 2007.

FYI I have attached a Word document with dChip output and settings
for the June 28 and the July 2 builds as well as LOH screendumps.

Hope you can fix this.

Best wishes,

Charlotte Schjerling, M.Sc.
RH Microarray Center


Posted: 13 Aug 2007 11:53 pm Post subject: difficulty in reading in
subarray CDF with latest dChip

--------------------------------------------------------------------------------

In this latest version of dChip for Affy 500k mapping when trying to
read in sty with subarray nsp, the program reads in both the sty and
nsp array as the first cdf. Thank you for your help.

Posted: 14 Aug 2007 01:58 pm Post subject:

--------------------------------------------------------------------------------

It's due to a new option to read SNP 6.0 generic CEL files. For other
arrays, make sure to check "Open group/options/Check CEL file type &
time tag".

Posted: 16 Aug 2007 09:36 am Post subject:

--------------------------------------------------------------------------------

Thanks. In my data set all files have the same time tag (actually,
there is no time tag in the header), so I have been routinely
unchecking that option. Any suggestions?

Thanks,
Diana

Please try this 8/16 version:

Http://biosun1.harvard.edu/~cli/dchip2006.exe

at "Open group/options", check "Check CEL file array type" but uncheck
"Check CEL file time tag".