 |
 |
 |
 |
I've accomplished what Linda wants to do as follows. I use a very
simple program I wrote called FCS2DAT to convert a list-mode FCS 1.0
file to an ASCII data file (columns of list-mode ASCII data). The
list-mode ASCII file can be read in by generic plotting packages for
making dot plots. Alternatively, the list-mode ASCII data file can be
processed by another program I wrote called HIST into histograms of
specified channel numbers (1024, 256, 128, 64, etc.). The lower
numbers of channels provide rather satisfactory smoothing. For
example, if 1024 resolution data are converted by HIST to 128
channels, adjacent groups of 8 channels are summed to make each of the
128 channels.
I've also developed a much more sophisticated program called MFI. For
your purposes, it has the advantage that it will produce GATED histogram
ASCII data files from ungated list-mode files.
My programs FCS2DAT.EXE, HIST.EXE, and MFI.EXE are available by
anonymous FTP from flowcyt.cyto.purdue.edu. For the former two, get
MARTZEXE.ARC and (if you don't have ARC.EXE) get ARCE.COM. For MFI,
get MFI.EXE and (if you use Hercules video) MSHERC.COM. You may also
want to get MARTZ.DOC. All 3 programs have built-in help. If you
haven't used FTP, I've enclosed instructions at the end of this
letter.
If any of my programs fail to work with your list-mode FCS 1.0 or 2.0
files, please FTP a sample file to Purdue and email me a message about
the problem. I'll try to adapt my programs to work with your files.
> Does anyone know if these files can be successfully imported and
> labelled etc using HG? Does it take FCS 1.0 format or is the ascii format
> necessary.
FCS 1.0 is a complex format which could not be read by generic plotting
packages.
> Is there a way to
> overlay plots from two different files using a graphics package after
> transfer?
No problem using the above strategy.
------------
Instructions for retrieving MFI.EXE (and MSHERC.COM if you use Hercules
monochrome graphics) from Purdue. These instructions are for a unix
OS computer -- details may vary for other systems.
At your internet-connected computer, issue the command
"ftp flowcyt.cyto.purdue.edu".
If the FTP command is functional on your computer, you will be asked
to logon. Give the name "anonymous". For a password, give your
complete email address. You will "arrive" at Purdue in the directory
pub. The command "pwd" (print working directory) will confirm this.
Issue the command "cd martz" to change to pub/martz. The command "ls"
produces a directory listing from which you can decide which files to
"get". Typically, issue the command "get mfi.exe" (and if you need
it, "get msherc.com"). You can see what else is available by getting
martz.doc, which also contains the dates of all upgrade postings.
Should you wish to upload any list mode data files so that I can try
to make MFI compatible with them, use the commands "cd .." (returning
you to /pub), "cd upload" (putting you at pub/upload), then "send
filename". (The file to be sent should be available in the local
directory from which you called ftp in the first place.) You may ARC
or ZIP files first if you wish.
When finished, say "bye" to ftp.
If you do not have access to an internet-linked computer which
supports FTP, ask your friends. This method is best because MFI is
improved frequently and by FTP you can always have the latest version
quickly. From our local unix computer, it takes less than 4 seconds
to get or put MFI.EXE between Mass. and Indiana! If FTP is not
accessible to you, I may be able to email you the program in uuencoded
form (be sure you have uudecode available before you ask me to do
this). This is more time consuming at my end, especially for overseas
mailings which may have to be split into 5 or so pieces to satisfy the
50K limit, then re-assembled by hand editing at your end. Within the
USA, it usually seems possible to email the uuencoded MFI.EXE in one
piece (about 210K). Snail mail of floppy diskettes is time consuming
at my end and by the time the program arrives it is usually obsolete.
-------------------------
--
/* - - - - - - - - - - - - - - - - - From - - - - - - - - - - - - - - - - - -
Eric Martz emartz@titan.ucc.umass.edu Professor of Immunology
Voice 413-545-2325, FAX 413-545-1578
Morrill IVN Rm 203, Dept Microbiol, Univ Mass, Amherst MA 01003 USA
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - */
|
|
|
|