The main objective for this release was to implement feedback, improve UX and optimize the 2.0 Genotyper application. With the release of Genotyper 2.1, the application is ready to replace the legacy application. The new algorithm is improved greatly and will be improved even more in the following release(s).
As many Users tested the 2.0 application, a lot of feedback was received. This feedback is very valuable to create an optimized solution. Based on the received feedback, multiple smaller changes were introduced. In the section ‘feature changes and improvements’ these changes are explained more in-depth.
- Parser improvements to support new or changed input files
- Authorization behaviour
- Switch to masterplates
- History visualization
- Team management
- Audit trails
- Algorithm improvements
- Updated “Last modified by” column
- Bug fixes
UI – UX
After an internal design review, the User Interface has been improved. These changes are very small but enhance the overall look and feel of the FastFinder Genotyper application.
Feature changes and improvements
In Genotyper 2.1, a user can only authorize an order once. After authorization, an export file is created that can be downloaded by the user. For Kraken users, results are being pushed back to the Kraken database. Also, after authorization, this data is used to train the algorithm and to populate the dashboards.
After authorization, it is still possible to change data and callings in the order. A user will not have the possibility to authorize again, but will be able to generate a new export file. By performing this action, the user can download a new export file with the changes included. For Kraken users, this action will push the changes to the Kraken database. This export will not be included in the training of the algorithm nor to populate the dashboards.
Switch to masterplates
In the PCR plate results table, the PCR plates are being shown by default. It is now possible to visualize the masterplates in this table as well. This can be done by clicking on the ‘switch to masterplates’ button in the ‘actions’ dropdown at the upper right of the screen.
As a user performs this action, masterplates will be shown for all assays in that order. A user can switch back the PCR plates by now clicking on the ‘switch to PCR plates’ button, or by refreshing the page.
An action toggle (eye icon) was added to the graph to show and hide the history model visualization.
In Genotyper 2.1, it will be possible to visualize the history for multiple readers at once. If a user selects multiple plates that were read with two different readers, both history models are shown. Different readers can have slightly different results for the same assay. A warning message will be shown to inform the user to be more cautious in calling the datapoints manually when multiple history models are shown. In the example below, two different history models are shown.
A fifth status was added as a possible calling: “Inconclusive”. In total, a user can give 5 different callings to a datapoint:
- Homozygous YY (red color): this calling is used if a user or the algorithm knows that the datapoint is homozygous YY
- Heterozygous XY (green color): this calling is used if a user or the algorithm knows that the datapoint is heterozygous XY
- Homozygous XX (blue color): this calling is used if a user or the algorithm knows that the datapoint is homozygous XX
- Uncalled (purple color): for datapoints about which the user or the algorithm is not sure how they need to be called
- Inconclusive (yellow color): for datapoints about which the user or the algorithm knows that they should never be called in that region
Important to understand, is that these inconclusive datapoints are also used to train the algorithm. This means that a user should also call datapoints inconclusive manually for every new assay in a region that should not be called. In total, the algorithm needs at least 5 inconclusive datapoints, coming from the same or multiple plates. When the algorithm is trained with the three clusters and 5 inconclusive datapoints, the history model will be complete. Of course, adding more training data will only enhance the history model.
In the side navigation of the Genotyper Application, the ‘Teams’ module is added. From this module, a user can manage access for team members. A user can either be an Authorizer or an Admin. An authorizer can edit and authorize orders. An Admin can perform the same actions as an authorizer and on top of that change general settings and instrument settings.
The audit logs module was added to the application and can be accessed by clicking the ‘Audit Logs’ button in the side navigation. In these audit logs, the following actions are being tracked per login user and time-stamped:
- Authentication log types:
- Successfully logged in
- Update log types:
- Datapoints for orderID ‘xxx’ and assayID ‘yyy’ were called manually
- Updated settings for instrument ‘xyz’
- Read log types:
- Successfully authorized order with Id ‘xxx’ and generated a CSV export
- Successfully regenerated CSV for order with id ‘xxx’
Please sign in to leave a comment.