The CDD Vault Visualization core application enables flexible reporting and dynamic analyses. All users can already add new data to an existing Visualization session, and display images and dose response plots in the table view. The new Visualization features to add some color to this Spring Season are:
Transpose the Data Table by Condition:You can now condense the display of aggregate data values (IC50s, averages, geomeans, etc.) by conditions in the Visualization report table. This creates one column for the calculation and another single column for the associated condition set.
Simply edit the table settings under the appearance tab and select the option to “Transpose protocol conditions”.
Then use the transposed column as parameters in plots (color, size, shape, axis) to facilitate data analysis.
Display Dose Response Legends:
Legends seen in the search report table are now viewable in a Visualization session as well. These legends are interactive. Hovering over a row in the legend highlights the curve in the mini plot view.
Excel Exports with Dose Response Plots:
Visualization table exports now have the option to include dose response plots. The plots are converted to static images embedded within the Excel file. Easily launch an interactive plot by clicking on the URL in the file next to the plot.
User-specified Zero-Click Inference Models:
To create a new model, users with sufficient edit permissions can now select any numerical readout definition directly from the Inference Models section of the protocol page. This only needs to be done once, and then the models will run for those readouts. The full list of options appears just above the model list. Checking a specific readout definition schedules it for training; once complete, it appears in the list of trained inference models.
Users can also control the visibility of released models. Click the eye icon next to a model's status to show or hide it. You can identify all hidden models using the quick filter checkbox.
In all cases, models are retrained automatically whenever new data are added to the protocol. The same release criteria apply throughout: only models meeting minimum performance thresholds will be released.
To provide feedback on features, or if you have questions and want to schedule a demo, please contact support@collaborativedrug.com.