March 14, 2014

CDD Vault Update: Averages and Normalization

Readout definitions: more flexible averaging

You can now choose a different level of aggregation for each average readout, whereas previously the aggregation level was uniformly set for all readout definitions in the vault.

Average_calculation_form

The available aggregation levels are as follows:

Batch/Run is good for replicate data in a single experiment (run). Replicated data from the same batch of a molecule will be averaged together for each run of the protocol.
Molecule/Protocol works well for generating summary results for molecules tested in a protocol. Data from all batches of a molecule will be averaged together across all runs of a protocol.

The calculation of averages is now a separate step: you have to create the readout definitions you want to average first, then click “add calculated readout definition”.

Calculated_readout_definition

Once you've configured an average readout definition, you will see it listed right next to the corresponding non-averaged definition on the protocol page (below), in search results, and on the run page's data tab. Note that you can not yet average normalized or dose-response readout definitions.

Complete_readout_definitions

Percent of negative control normalization for dose-response calculations

While positive controls are almost always a good idea, some assays are run without them. Until now, dose-response calculations required both a positive and a negative control if the raw data was to be automatically normalized within the CDD Vault. In this release you can normalize your dose-response data as percent of negative control. You will see the new option when you define a dose-response calculation in your protocol:

Crc_normalization

Molecule description has moved

The “description” field, which was previously part of the core molecule definition, is now just another user-defined field. If you had any data in the description field, it was automatically migrated.

Description_move