TEST Auto Mapper / Mapping Suggestion

Automate the mapping of your experimental data sets

Written by Eduardo Abeliuk
Updated over a week ago

The TEST application has a powerful and flexible data model specifically designed to store experimental data. Each assay (a.k.a., experimental run) that is recorded in the TEST application can be loaded with experimental data tables associated with a particular experiment.

The users can either manually annotate the different datasets for column types or use our auto mapper feature which relies on machine learning capabilities to automatically annotate the data with their relevant column types.

Here we are going to learn how to execute the latter i.e. the automatically annotating your datasets with column types.

Click on Assays from the top navigation bar in the TEST application.

It will take you to Experimental Data interface from where you can either access your pre-registered experiments or import new (assay) files. In this example we are going to import a new file.

This will open up an import file dialog box where you can upload your CSV file, and name your Assay as relevant. Once done click on Start.

(Side note: There is an additional field that allows you to select a parser. The parsers allow you to avoid performing the following steps several times, provided they have been done previously.)

This will take you to the Data Formatting interface where you can restructure and reformat data as per your experimental needs. Click on Next at the bottom of the interface and move to the Data Mapping interface.

On the Data Mapping Interface, you will see undefined classes and sub-classes. Instead of going through each individual column and naming it with a descriptor as per the data type, you can use the Mapping suggestion option by clicking on mapping presets.

You can now see the Mapping Suggestion dialog box. This will identify available data types in your experimental data and then will automatically map your columns with relevant class and sub-class types, using machine learning algorithms.

You can either check the Accept all box or individually check the box for each row to confirm whether the mapping suggestions fit your needs, and once did click the confirm button to "auto" map your Assay datasets.

This will also enable the validate button on the bottom right of your screen meaning that your mapping is correct. Once validated, a Save option will replace Validate.

You can then see your Assay at the main Experimental data page on the TEST application and access it from there.

Supplementary Readings

Did this answer your question?