This release brings many sought-after enhancements for dose-response curve fitting, such as overriding IC50s and constraining fit parameters for individual curves. Plus much more; keep reading to find out! Amidst all of these features for biologists, we’ve added one new feature for chemists especially: preserving molecule orientation as drawn.
Chemists get just one feature in this release, but it’s a good one! Importing chemical structures from Molfiles will now save and render all of the detail captured by this representation, including annotation and orientation.
Where possible, we are recovering the Molfiles used to import existing molecules, so within a day or two of this release you may see all or most of your structures rendered exactly as you imported them.
A note of reassurance: SMILES format will continue to be supported without changes, as will structure validation on import for all structures: learn more.
If you perform assays that measure growth rather than inhibition, the values and curves that you see in CDD Vault are probably the inverse of what you want. Well, now you can create the correct calculation, rather than standing on your head when looking at the plots. In addition to normalizing raw data to calculate percent of control or percent inhibition, you can calculate “100% − ” those values. This normalization option is available when defining a normalized readout for single point data or dose-response calculations.
The example screenshot below shows a section of the dose-response setup, but the form looks very similar in both cases.
For dose-response data, this will essentially flip the curve.
You now have the option to override a calculated dose-response value when the curve fit produces a mathematically correct, but biologically invalid result. Maybe this is a feature for chemists after all: it will help them interpret those especially badly behaved curves.
A calculated value may be manually changed to:
The override feature is available via the dose-response mini-app, which opens up in a new window when you click on the plot in the search results table, or on the “Flag outliers and Override” link right below it.
Don’t forget to save your changes!
Values that have been overridden will appear italicized and underlined in the search results— this formatting will be familiar if you’ve removed outliers from curves in the past. Also, such values are easy to find using the protocol search options.
When setting-up a dose-response calculation, you already had the option to lock individual fit parameters uniformly across all curves in the protocol. Now you can restrict parameters to a range, but allow them to float within that range.
The new options are available in the same section of the dose-response readout definition. You can define an explicit range (“from”), or a lower (“≥”) or upper (“≤”) bound of each of the three curve fit parameters, in addition to the existing exact number (“=”) option.
For example, if you would like all curves in your inhibition assay to have a positive slope, you can set the Hill slope to be ≥ 0.
The curve fitter will optimize the parameters as usual, except it will enforce the ranges you have specified.
With a well-defined set of protocol-level parameters constraints, the majority of your curves should fit correctly. However, if you need to refine parameter constraints further, you may do so for each individual curve.
This feature is found in the dose-response mini-app, together with the calculated value override.
If protocol-level parameter constraints are present, they will be automatically populated for each curve. You may then override these defaults with a new set of parameters, or choose to float parameters by clearing the values: available options mirror the options available at the protocol level.
You can now set the minimum activity threshold, so that bogus IC50s will not be calculated for compounds that show insufficient activity. Instead, the reported value for these curves will be “> highest tested concentration”. This setting is part of the dose-response readout definition, and has two options:
After you have gone through your curve validation and QC steps, you can check the statistical confidence of the calculated values.
The upper and lower limits of the 95% Confidence Interval are provided together with the rest of the curve statistics. You can filter out protocol results by the upper and lower 95% CI limits:
Then customize your report options to view them alongside the IC50 values:
How’s that for confidence!
Please refer to the dose-response toolkit on our help site to learn about setting up dose-response calculations in your vault.
We have added a vault setting to prevent unintentionally sharing chemical resources between projects when importing data. Many of you work with multiple collaborators where accidental sharing of intellectual property has grave consequences, so we have turned this on by default. The import validation report will show sharing as a suspicious event: the user must explicitly accept the event.
Previously, CDD would bring the sharing to your attention as a noteworthy event, but no explicit acceptance was required. If you prefer this, vault administrators can update the setting.
When this box is not checked, sharing is automatic, and a noteworthy event serves as an alert that molecules/batches or plates are being shared between projects. If the sharing event is unintentional, it may be overridden using the “reject” option.
The title here describes it well: when you download a detailed excel report with many images, the exported file is smaller in size and takes less time to download.
We’ve tidied up the formatting of all calculated chemical properties to make them consistent throughout the website and in file exports.