In Dr. McCrea's lab here at the SCRC, we've developed a system of tracking
all our experiments in a searchable database.
This system involves a systematic approach to organizing and naming the
experiment data files and directories, as well as paying close attention
to the run descriptions, which we edit after the experiment to assure a
good set of searchable keywords.
The whole procedure is explained in our on-line tuturial #10,
How to update, backup and insert new database records.
A key part of this procedure is the transfer of experiment data file summaries
from the Linux capture and analysis system into a FileMaker Pro database.
This is accomplished with a shell script we developed specifically for this
purpose, the
getrundata program,
which you can download from our
analysis script archive.
This program generates CSV (comma-separated values) records, to be imported
into the database, which summarize the experiment name, run name, path to
the file, length, and channel names for each file, as well as the run
descriptions.
Once imported into specific fields into the database, many other pieces
of information can also be edited to aid with searches.
Another feature of this database is it includes a link for each run file
in the database, to a summary page that is generated by a web server running
on our analysis system.
These summary pages show a thumbnail image of the data in that run, which
appears something like this:
Also featured on the actual summary pages, which require a login to the server,
is a link to fire up the analysis program on that run file, using a
remote X11 session.
This allows one to search for data that fits specific criteria, quickly
view summaries of these experiments to pick out the ones of interest,
and then go on to analyze the runs you've picked out, all from the PC
at your desk.