Before you can access the tool you’ll need to be granted access. Once your access is granted, you will receive an email with instructions on how to setup your password and complete your Blackfynn profile. Once you have created your password you can access the PPMI Progression Analysis Tool by visiting:
Use the email address and password that you used to create your account on Blackfynn to access the Progression Tool.
The Progression Tool Dashboard
Once you are signed in to the tool you will be presented with a dashboard powered by PPMI data. The dashboard is a combination of filtering tools and data visualizations that enable you to explore and analyze PPMI data.
At the top of the page you’ll notice a slim bar containing containing menu items for:
- Severity Score
Each of these menu items contains tools to help you filter and set parameters according to your desired definitions for disease severity and progression. Each menu contains a “What’s this?” link that contains helpful hints for each section.
Below the filters you will see a set of data visualizations that give you an at-a-glance view of your:
- Total Subject Pool
- Age Distribution at Baseline
- Sex Distribution
- Diagnosis Counts for Subjects in the Pool
The additional interactive charts display the PPMI data in a variety of ways to help aid you in your research. Current charts include:
- Mean Measure Over Time: Displays the mean value of the selected variable over all visits for the provided subjects (after initial filtering). Only patients/visits where the variable is assessed are shown. Here, we represent Baseline as month 0, and Screening as month -1. Note that we do not plot the visits for which the number of subjects is less than 10% of the total number of subjects in the pool. The default variable is Severity, as defined in the dashboard.
- Measure at Baseline: The histogram displays the distribution of the selected variable, used in the Mean Measure Over Time chart, for all participants at baseline.
- Disease Progression: Relationship of specified baseline feature (y-axis) with disease progression as defined in the dashboard (x-axis).
- Distribution of Disease Progression (Optional): This boxplot displays the distribution of the selected variable for each group. This chart is only displayed if you have selected a grouping option from the Grouping menu.
- Baseline Features Predictive of Progression: This table displays the top 10 most important baseline features for predicting progression based on the Gini importance.
Each of these charts displayed on the page will update automatically as you filter your data.
Some charts provide additional ways to explore your data. Charts that include blue text in the chart header can allow you to explore other views of your data or variables by clicking the blue text and selecting another option from the menu.
Filtering the Data
The Subjects menu allows you to filter out PPMI subjects by selecting one or more filtering criteria. By applying one or more filters you are reducing your subject pool for the Progression Tool.
You can add a new subject filter by clicking the blue plus (+) sign at the bottom of the menu. After clicking the plus sign a new row will appear that will allow you to select an additional property to further filter. Clicking the Property dropdown will display a list of properties you can choose from. After assigning a property and value to filter by, click the Apply button to update your subject pool. The data displayed in your dashboard will begin to update once you apply your changes.
Each filter is applied in combination such that each filter must be true for all individuals. Note that all filters are applied to subjects at baseline — filtering based on other timepoints is not possible.
To remove a subject filter, simply click the “X” located at the end of the property you would like to remove.
The Severity Score filter allows you to determine which metric you will use to determine severity of progression.
Clicking the dropdown box will display a list of metrics from you to choose from. You can select additional metrics to create a “composite” score for disease severity by clicking the blue plus sign. The ranges/directionality of each assessment will be taken into account when calculating a composite measure. If only one measure is selected (default), then the raw scores will indicate severity. When multiple severity variables are defined, the features are rescaled as follows:
For increasing severity scales, such as UPDRS:
x’ = [ x - min(x) ] / [ max(x) - min(x) ]
For decreasing severity scales, such as MoCA:
x’ = 1 - [ x - min(x) ] / [ max(x) - min(x) ]
Additional metrics can be removed by clicking the “X” to the right of the metric dropdown. You are required to identify at least one metric.
Clicking the Apply button will update the data on your dashboard.
Disease progression is defined by the change in severity between the selected time span.
You can change the time span by clicking and dragging on the white circles on either side of the line. As you drag, the handles should snap in place and display the year. The data in the dashboard updates automatically as you drag the progression timeline handles. Progression is computed as a simple subtraction of the severity scores between the later and the earlier selected times.
The Grouping menu allows you to group your data using the provided grouping options:
- No Grouping: this is the default grouping. The data is displayed as a single cohort.
- Fast vs. Slow: divides the data into two groups based on the computed progression values (fast progressors vs. slow progressors).
If you choose to group your data, the dashboard will update automatically to reflect your choices. An additional chart will be displayed alongside the Disease Progression to show the distribution of your chosen measure.
You can export your filtered data at any time by clicking on the “Export Filtered Data” button located in the blue bar at the top of the page.
The data will be exported as a CSV file and will contain the participants whose characteristics match those defined in the Subject Filters.
Most filters and charts contain a “What’s this?” link. Clicking the link will display a small pop-up containing additional information about each chart or filter.
If you have any questions or need help, send an email to email@example.com. Someone from our team will reply to you within one business day.