Skip to content

Getting started

App

The AutoIMS app will generate configuration file which can be understood by our AutoIMS-Runner app in Delft.

The runner app works by continuosly syncing data from Vanderbilt to Delft and executing executing tasks that have been specified by the user (e.g. you). The way you specify the tasks is very simple. You select your project (see below), select tasks from the Available tasks list, modify any parameters you wish to modify, add the task to the Selected tasks list and then click Save configuration. Once that is done, that configuration file alongside all the data is copied to Delft, and we run the workflow.

Folder structure

All data must live within the DELFT folder, and then within you own folder.

DELFT
├── lgmigas
    └── project-1
        ├── config.json
        ├── extras
        │   └── masks
        │       ├── dataset_1_mask.geojson
        │       ├── dataset_1_registration.i2r.json
        │       ├── dataset_2_mask.geojson
        │       └── dataset_2_registration.i2r.json 
        ├── raw
        │   ├── dataset_1.d
        │   ├── dataset_2.d
        │   └── dataset_X.d
        └── tags.csv

Meaning of each folder/file

  • The DELFT folder is divided based on user names so that everyone has their own space for experiments.
  • Inside each user folder, you can create projects that would encompass different types of analyses.
  • In the raw folder, you can put your Bruker (.d) or Waters (.raw) datasets.
  • In the extras folder, you can provide additional files such as masks, databases, README, etc.
  • In the extras/masks folder, you can provide GeoJSON masks and i2r.json registration files (from image2image).
  • In the extras/databases folder, you can provide CSV files with database(s) to be used by annotine.
  • In the extras/peaklists folder, you can provide CSV peaklist files to be used in centroid data extraction.

Inside each user folder, you should create projects that would encompass different types of analysis.

Structuring your own data

Using Windows Explorer/Finder

Here are some basic steps: 1. Inside the DELFT folder, please create your directory (e.g. for me, this would be lgmigas). 2. Inside that folder, add the name of your project (e.g. kidney_matrix_pos) 3. Inside that folder, you create folder raw where you would paste your raw data (e.g. Bruker .d) 4. Then you can use the AutoIMS App to generate your configuration.

Using the app

  1. Open the AutoIMS app.
  2. Click on the Select project directory button.
  3. Select directory within the DELFT folder.
  4. Create your new project folder (e.g. kidney_matrix_pos) as above.
  5. Click on the Create template directory... button.
  6. TA-DA! The folder was pre-populated with a few files.

Refeence Project

AutoIMS permits creating links to other projects which enable reusing of already existing data and potentially creating new visualisations or analyses. This can be very useful when working on a long-spanning project where data is continuously acquired, however, certain analyses can be performed on the incomplete data (e.g. HuBMAP or KPMP).

There are some limitations to what tasks can be performed on the reference projects. Essentially, none of the pre-processing tasks will be executed (e.g. data conversion, m/z or IM alignment, m/z calibration) but many others are permitted.