Goal:

The initial goal of the microarray component of the Systems Biology
Experiment Analysis Management System (SBEAMS) is
to streamline the process of organizing the information
associated with requesting, producing, and processing microarray
experiments, so that MIAME/MAML compliant datasets are produced for
subsequent analysis.  These datasets may then be exported to be used with
analysis packages that (will) work with MAML data, or exported into any of
several external formats to expression analysis. In phase 2 of the
project, expression analysis software will be integrated with the database
and interface itself.

------

Overview of the process:

Requestor:
 1 - Fill out a Project entry unless continuation of existing project
 2 - Fill out array request form and submit

Data producer performs the following for each slide in the project:
 3 - Insure all required information is up-to-date:
     - slide model, slide lot, slide
     - protocols -> hardware, software
     - array layout information
 4 - Perform and enter information about the printing
 5 - Perform and enter information about
   - receipt of samples
   - labeling
 6 - Perform and enter information about hybridization batch
 7 - Associate (project, layout, printing, slide #, hyb) -> "array"

 8 - Perform and enter information about the scan
     Data products are staged in:
       /net/arrays/data/$array_number/scan01/   (TIFFs, etc.)
 9 - Perform and enter information about the quantitation
     Data products are staged in:
       /net/arrays/data/$array_number/scan01/quant01/ (dapple or QA files)

When some (and/or all) slides in a given project are quantitated:
10 - Database produces best guess processing plan (for mergeReps, etc.)
       Requestor or Producer reviews processing plan and submits
11 - Batch job is queued to run preprocess, mergeReps, SAM, VERA, mergeConds
     Input staged data is archived to:
       /net/db/sbeams/microarray/project/$project_id/arraydata/$array_number/etc.
     Pipeline data products are stored:
       /net/db/sbeams/microarray/project/$project_id/proc01/proc.log & data
     Requestor receives processing errors/confirmation


Comments:

- Data processing pipeline steps can be readjusted and rerun
- Resulting data products are maintained in native format
- Meta-data for production and processing steps are stored in the database
- Certain data products are ingested into database for further processing:
  - Layout (key) data
  - Raw spot measurements for each slide
  - Expression ratio measures for each "merged condition-element"