Study day, visit number, and visit

Test notes

• A checkbox appears beneath the x-axis dropdown labeled Inliers that toggles the display of participants in each chart without values outside the normal range.

Consider using alpha blending (like in safety-histogram)

preferably via defineStyles()

... with cool interactivity like this to define the baseline value.

all css should have a #safety-outlier-explorer selector.

When provided, pass time_cols[].order to the webcharts settings as x.order.

Issue was originally: "Highlighted listing displays differently than un-highlighted listing". New title represents reduced scope (as described in comments below).

Currently can't see any data on a panel with a single time point

Test notes

• Edit ADBDS.csv so that at least one test has only one timepoint. For example, filter the data on Albumin, delete all records except Visit 4, unfilter the data, and upload it to CAT. That test should have enlarged marks.
  • Open CAT.
  • Select safetyData/ADBDS.csv in the dataset dropdown.
  • Click the magnifying glass to pull up the data preview.
  • Export the data to CSV.
  • Open the export.
  • Filter to a single TEST.
  • Filter on all visits except one.
  • Delete all records.
  • Remove the filters.
  • Upload the data to CAT by clicking Choose File and then clicking Load. The edited data file should now appear in the data file dropdown.

Request from Jack - hours and days are smooshing together for a client's version of the PoE and we would like to fix this for him ASAP. Projected deadline is mid-Jan

Test notes

• A checkbox appears in the Controls section that toggles the display of visits without data.

A new label is generated each time the panel is rendered with 0 records.

Use new version of webcharts for pagination - remove functionality from paneled-outlier-explorer repo.

they are fairly common in units in our sample data

Refactor the data listing view so that each person gets a section (tbody? div?) that has a summary row (subjectid, demographics, etc) and then has one row per test, with columns for visits.

David Hall mentioned the possibility of using the Paneled Outlier Explorer with big data in the future and was curious if we could predict how it would perform with a much larger data set.

We should assess the performance to see if there are any updates that would result in quicker load times and a more responsive tool.

David said that this is a low priority for now.

One idea is to see if the number of marks that are displayed is causing it to load slower.

Test notes

• Setting time_cols[].order_col to VISITNUM orders the x-axis domain chronologically.
• Not specifying time_cols[].order_col the x-axis domain alphabetically.

Test notes

• Set the x-axis to an ordinal variable, e.g. visit.
• Brush a chart, maximize it, minimize it, hide it, bop it, shake it, make sure the brush doesn't screw up.

To avoid overlap like shown:

Bombs at this line if no records are found for a given panel.

Currently the data listing appears and functions correctly but leaves this large white space:

Have each option start on a new line:

Similar to the remove button.

Right now they are being removed without comment here: https://github.com/RhoInc/paneled-outlier-explorer/pull/19/files#diff-f2d1bf62fd73c454622443228f656928R16

e.g. in onResize():

        if (this.filtered_data.length && this.filtered_data[0][this.config.lln_col] && this.filtered_data[0][this.config.uln_col])

Test notes

• Edit ADBDS.csv so that at least one test has no normal range values. For example, filter the data on Albumin, delete the normal range values, unfilter the data, and upload it to CAT. That test should not have a normal range when the Normal range dropdown is set to LLN-ULN.
  • Open CAT.
  • Select safetyData/ADBDS.csv in the dataset dropdown.
  • Click the magnifying glass to pull up the data preview.
  • Export the data to CSV.
  • Open the export.
  • Filter to a single TEST.
  • Delete all values in STNRLO and STNRHI, the normal range variables.
  • Remove the filter.
  • Upload the data to CAT by clicking Choose File and then clicking Load. The edited data file should now appear in the data file dropdown.
• Input a bad normal range variable name and verify that the Normal range dropdown does not include LLN-ULN.

There is something special about the AST chart that makes it displays slightly offset in Internet Explorer only. I have unchecked several of the other charts to shift them around and it appears that it has nothing to do with where it is on the page; it seems to be specific to the AST chart. Please see the image below.

Current settings make many panels impossible to read: