ExpressionData:ArrayExpressSubmission
From VectorBase Help System
Contents |
[edit] Prerequisites
Your array design and expression data must already be loaded into VectorBase BASE.
[edit] Step 1: Deposit array design at ArrayExpress
Follow this link to submit your array design to ArrayExpress, where you will require the following:
- An ADF file [1].
- If you submitted a .gal file to us, we can generate an ADF file for you directly from our database. Please skip to point 2 (protocol text).
- You can create ADF files from several array design formats using a tool [2] at the EBI (make sure you use the same array design which was submitted to VectorBase).
- Do not spend time producing elaborate database annotations for your ADF file. You only need one column with the heading "Reporter BioSequence DatabaseEntry [vectorbase_reporter]" (((TO BE CONFIRMED))) which should simply contain Reporter Identifiers.
- You do need to spend some time providing the correct information for the following columns in the ADF file (see [3] for more information), unless you already provided us with this information during array layout submission (which, ideally, you would have done).
- Reporter BioSequence Type as one of
- cDNA_clone
- ds_oligo
- genomic_DNA
- PCR_amplicon
- ss_oligo
- Reporter BioSequence Polymer Type as one of
- DNA
- RNA
- Protein
- Reporter Group [role] as one of
- Experimental
- Control
- Reporter Control Type as one of
- control_biosequence
- control_buffer
- control_empty
- control_genomic_DNA
- control_label
- control_reporter_size
- control_spike_calibration
- control_unknown_type
- Protocol text for your array design and manufacture
Make a note of the ArrayExpress accession number assigned to your array design.
[edit] Step 2: Finalize sample annotations and experimental factors
Your sample annotations in BASE should follow the MGED ontology [4] if at all possible. You may need to email us to provide new sample annotation types and options for them. Note that there is a bulk sample annotation facility in BASE.
Make a note of which sample annotations are experimental factors (i.e. the conditions you tested in the experiment).
[edit] Step 3: Create a BASE experiment
If you haven't done this already, create an experiment (under the Analyze data→Experiments menu item) and add all the relevant raw data sets to this experiment (using the Analyze data→Raw data sets menu item).
Fill in as much information as possible in the Experiment's info tab. The description text should describe the whole experiment (this will go into ArrayExpress).
[edit] Step 4: Prepare protocols
If your protocols are not already in ArrayExpress, write a detailed paragraph to explain each of the following steps: grow, treatment, extraction, labeling, hybridization, scanning, image_analysis.
If your protocols are already in ArrayExpress, make a note of their accession numbers.
[edit] Step 5: Final data matrix
To be fully MIAME compliant, ArrayExpress require the submission of processed data which you have refered to in your publication - this is usually filtered (no bad data), normalised and averaged over replicates. See this page for more details of the simple file format used.
Please also write a short piece of protocol text describing the data processing you performed.
[edit] Step 6: Authorize submission
Send us an email containing all the information requested above (that is not already in BASE) and we will submit your experiment to ArrayExpress using the TAB2MAGE program.
