Assay annotations play a critical role in drug discovery to ensure consistency and reproducibility of experimental results, interoperability of data sets, semantic consistency (standardized ontology templates), and readying data sets for AI and ML workflows. Describing assays using ontological realism provides a mechanism for capturing data provenance as well as facilitating the ability to integrate, search and compare the results of different studies and investigations 1.
CDD Vault offers an assay informatics tool kit consisting of custom protocol fields, custom run fields, standardized out-of-the-box ontology templates, and search/aggregation tools such as the hierarchical aggregate protocol property grid and ontology filters.
Custom protocol fields (top-level), in combination with a standardized ontology template, will capture consistent metadata for an assay while custom run fields (second tier) will capture metadata specific to a run of an experiment. Most likely, run fields will vary from experiment to experiment while protocol level fields will capture metadata relevant to all runs. This two-level annotation schematic allows for users to capture all relevant metadata side-by-side with their experimental data readouts.
Protocol Fields
Vault Administrators are responsible for creating any number of top-level protocol fields that can be used to fully describe an assay. These fields are Vault specific and may be optional, required, or conditionally required. These values are fixed for all runs of an assay. Any metadata that doesn’t change over time can be captured in the Protocol definition in a similarly structured manner.
By default, all Protocol fields will be immediately available for every Protocol definition as soon as they are defined. The use of Protocols Forms allows further organization of meta data fields that are configured with specific subsets of fields relevant to a general category of assay.
As shown in the simplified example below, Protocol Fields can be populated when creating a new protocol…
…and can be edited on the protocol overview page under the protocol definitions section.
Helpful Hint: Don’t rely on a verbose “Name” to convey all information. Use Protocol Fields to capture the necessary details and use a shorter, more concise Name for your Protocols. Using protocol fields also unlocks powerful search functionality.
For Vault Administrators who may need a refresher on how to configure Protocol Fields, please review this article.
Run Fields
Complementing Protocol Fields and Readout definitions, Run Fields are used to annotate each individual experiment. These values can change over time but will be consistent for all data imported to a single Run. These fields will be shown anytime a run is created or edited and can be displayed within the search results table by customizing the report.
Run field annotations can be modified and/or expanded to include custom Run fields with defined business rules such as CRO, Assay Kit, Assay Target, etc. The Vault Administrator must create these new fields that users can then populate when creating new Runs either through the Run Data interface of a Protocol……
…or through the bulk importer.
Please navigate to the Protocols Forms section of this article to configure specific subsets of run fields relevant to a general category of assay.
Ontology Templates
Users can also include hierarchical ontology templates in Protocol and Run field definitions. Annotation of bioassay protocols using semantic web vocabulary is a way to make experiment descriptions machine-readable. Protocols are communicated using concise scientific English, which precludes most kinds of analysis by software algorithms. Given the availability of a sufficiently expressive ontology, some or all of the pertinent information can be captured by asserting a series of facts, expressed as semantic web triples; subject, predicate, object. With appropriate annotation, assays can be searched, clustered, tagged and evaluated in a multitude of ways, analogous to other segments of drug discovery informatics.
CDD Vault offers two standard ontology templates to work from including the Common Assay Template and the Pistoia Alliance Assay Template. Protocol Forms allow you to pick and choose which fields from the ontology are relevant for each category of assay.
Common Assay Template
The CDD Vault Informatics Team curated the Common Assay Template (CAT), which is designed to capture the most high value metadata relevant to bioassay protocols. This template draws from a number of underlying vocabularies created by domain experts, including the BioAssay Ontology (BAO), Drug Target Ontology (DTO), Cell Line Ontology (CLO) and others.
Due to the complexity of traditional ontology template use-cases, scientists cannot be reasonably expected to effectively leverage hierarchical ontologies without having access to software tools that make it straightforward to use the vocabulary in a canonical way. CDD Informatics sought to remove this barrier by:
- Defining a BioAssay Template (BAT) data model
- Creating a software tool for experts to create or modify templates to suit their needs; and
- Designing a CAT to leverage the most value from the BAO terms2.
The CAT was carefully assembled by biologists in order to find a balance between the maximum amount of information captured vs. low degrees of freedom in order to keep the user experience as simple as possible.
Vault Administrators will be able to navigate to Settings -> Vault -> Ontology Templates to view and edit the Common Assay Template.
Pistoia Alliance Assay Template
The Pistoia Alliance is a global, not-for-profit alliance of life science companies, vendors, publishers, and academic groups that work together to lower barriers to innovation in R&D.
Their projects transform R&D innovation through pre-competitive collaboration by bringing together key constituents to identify the root causes that lead to R&D inefficiencies. The Pistoia Alliance develops best practices and technology pilots to overcome common obstacles where members collaborate as equals on open projects that generate significant value for the worldwide life sciences community.
The Ontologies Mapping project was set up to create better mapping tools and services, and to establish best practices for ontology management in the Life Sciences. Ontologies can include hierarchical relationships, taxonomies, classifications and/or vocabularies which are becoming increasingly important for support of research and development.
Vault Administrators will be able to navigate to Settings -> Vault -> Ontology Templates to view and edit the Pistoia Alliance Assay Template.
Using Ontology Templates:
In this example we will create and annotate a GABA-A radioligand binding affinity protocol using both custom protocol and run fields as well as the Common Assay Template ontology.
After filling out any custom protocol fields, users can begin to annotate this protocol using the standardized hierarchical ontology template.
Users will have the option to fill in the ontology template fields either with an “annotation” or “free text”. Free text will bypass the standardization so it is recommended that this option is only used when no appropriate annotation exists in the template. Lets start with the first field “Bioassay Type”. By selecting “Annotation”, a popup will appear with this segment of the hierarchy. We can select “binding type” given that this applies to the protocol we are creating.
Next, we will annotate the “Bioassay” field to specify what kind of binding experiment/ process is occurring. In the below screenshot, we have searched for “radioligand” and can see that it exists generally under the “Pharmacodynamic Assay” branch in “Primary Pharmacodynamics” as a binding assay type.
Given the extensive granularity of the ontology templates, users may consider creating protocol forms where they can pick and choose which fields are relevant to which experiment types. In the below screenshot, only a handful of fields were annotated - creating a protocol form with only these fields will streamline future protocol creation and protocol organization.
Protocol Forms
Often, Protocol and Run Fields will only apply to a subset of your Protocols given that different experiment types are associated with different descriptors. For example, the fields used to annotate an in vitro radioligand binding affinity study will be significantly different than those of an in vivo toxicology study.
To streamline the protocol creation process, users can leverage Protocol Forms to define a concise set of relevant fields for that general category of assay (animal, biochemical, biophysical, cell, enzyme, etc.). Forms can be selected when creating new Protocols or editing existing ones, and users can even update forms over time without losing data since the fields are stored for every Protocol; the Form simply dictates which fields are displayed.
A single Protocol Form will control which fields are shown on the Protocol and Run definitions since these are correlated.
Creating a new protocol form
Vault Administrators may create new and manage active protocol forms by navigating to Settings -> Vault -> Protocol forms interface. To create a new Protocol form, select the “Create a new form” button in the top right of the settings menu.
As shown in the image below, the following pop-up interface will have two tabs; Protocol Definitions and Run Definitions (1.). Relevant fields will first be selected for the Protocol level and then the run level. Each Protocol Form will be given a name (2.). Users may consider creating a form for each general experiment type that they will be conducting such as a form for each in vitro, in vivo, and in silico experiments. Of course, this is the most general categorization and some groups will need to create more granular forms to properly annotate their experiments.
As described above, custom user-defined protocol fields (3.) will be available within the forms editor. To bring a subset of protocol fields into the form, drag and drop them from the right-hand menu into the free space. You may include up to four fields per line and adjust the order of fields by dragging them to a new position. Lastly, consider selecting a subset of fields from a standard ontology template (4.) to use in conjunction with your custom fields. For those users who only want to use the standard ontology fields, create a form using only the template fields or select the ontology template of choice when creating a new protocol to annotate that protocol with the full ontology hierarchy.
Once a field has been added to the form, use the pencil icon to:
- define business rules (optional or required)
- provide default values
- lock default values
- configure allowed pick list values
- Select default values from ontology templates
- Define allowed ontology branches
The following screenshot shows the appropriate protocol form matching the relevant information described in the GABA-A radioligand binding assay example described above.
A similar workflow will be completed for the run fields relevant to this protocol form. Users will have the option to choose from their own custom run fields and any of the standard ontology template fields as well.
Helpful Hints:
- Optional Protocol fields can be made required for specific forms
- Provide default values to speed up data capture
- Lock values for specific fields thus providing more detail about that type of assay without creating more work for the person defining the Protocol
The implementation of Protocol and Run Forms within your assay informatics strategy can:
- Bring assay awareness to each scientist so key details are captured
- Ensure assays are fully and consistently annotated so data isn’t lost over time
- Shorten Protocol creation time by providing specific fields instead of generic text boxes
- Elucidate which assays have been performed and how they relate to each other, key for QSAR/AI models
- Share assay details with collaborators, CROs, publications, etc. using a common universal vocabulary
-
Make assay data precisely searchable: find protocols with specific characteristics, either using specific terms or branch categories within the ontology
Property Grid and Searching
Providing assay annotations allows for another layer of searching and aggregating data within CDD Vault.
Under the Protocols tab of the Explore Data page filters can be created to search across all assay annotations. Select “add a new filter layer” and then choose from any custom Protocol Fields or branches of the ontology template(s) used within the Vault.
For example, we may select “Bioassay” from the Common Assay Template and then choose to search by the most general branch of the hierarchy (Pharmacodynamic Assay) or choose to search by any of the more granular branches all the way down to the type of binding assay = Radioligand. If a user selects to search by the most general branch, the search results will return all protocols that have been tagged with values equal to or under that particular branch of the ontology tree. Notice that branches with tagged values will be bolded and have a descendant count next to the field name.
Once a search has been executed, consider clicking on “Explore Protocol Data” to aggregate together all of the molecules that have been tested across the returned assays.
For a visual representation of the assay annotations, navigate to the “Property Grid” subtab of the Protocols tab.
Protocol names will be displayed across the top column headers while Protocol Fields (including ontological terms) will be displayed in a hierarchical manner across the lefthand-side rows. Half filled squares represent that a more specific term was used to annotate that assay while a fully-filled square represents the exact term used to describe that assay.
Specifically for custom protocol fields, consider using the Protocol Field search mechanism on the traditional search page to aggregate data based on custom annotations as opposed to specific protocol searches.
For additional information or training, please contact the CDD Support team (support@collaborativedrug.com).