Processing math: 100%

Welcome in SHADE


Brief overview

SHADE is a Shiny application for providing support to researchers in design of experiments. It includes a tool to perform power analysis for usual test statistics, as well as a reporter tool to draft the ethics committee form for animal experimentation (CETEA), in a user-friendly interface. Initially created to design experiments using animals, SHADE might be used to design any experiments (cell culture…).

The global workflow of SHADE is detailed hereafter:

What's new in SHADE

Upcoming release - SHADE version 4.0

Future version is being developped to improve user interface and facilitate navigation


October 2023 - New version of SHADE is online !

User interface improved with an advanced tab for exploring parameters and language option in final report. Cage effect visualisation was also improved.


February 2021 - Version 2.0 released

User interface was improved with validation buttons, tabs for exploring parameters. Bugs were fixed such as external data loading and numerical errors in power analysis.


March 2019 - SHADE is online

Implementing power analysis for frequently used statistical tests, power graph and a report for CETEA form.


Define parameters

First, choose a test family

Save for next steps and send to report

Visual example

Run power analysis


Choose test parameters

0.10.050.0010.0410.081
0.510.80.50.70.9
0.01320.011.212.41
10.2500.40.8
15415913

Save for next steps and send to report

Estimate effect-size with relevant previous experiment

Import data

Save for next steps and send to report

Overview of data format

Example of expected input :
(if no picture appears above you should disable your addblocker for this page)
Your imported file :

What is effect size ?

Here is the definition of Cohen's h :
Effect size for comparing two proportions

Below is a toy example to illustrate effect size for comparing two proportions, using Type one error α=5% and a power of 80% :

0.001320.0011.22.4

What is effect size ?

Here is the definition of Pearson's correlation coefficient :
Effect size for comparing two continuous variables

Below is a toy example to illustrate effect size for comparing two continuous variables, using Type one error α=5% and a power of 80% :

00.500.20.40.8

What is effect size ?

Here is the definition of Cohen's d :

Effect size can be interpreted as the number of standard deviation between the means of the two groups.

Below is a toy example to illustrate effect size for a paired comparison, using Type one error α=5% and a power of 80% :

0.001320.0011.22.4

Make a CETEA report (DAP form) and export it

Specify features of your experiment

Cage

To contact us

Click here to contact the SHADE authors by email.

SHADE is an application developed by members of the Hub of Bioinformatics and Biostatistics of the Institut Pasteur.

It provides support to Pasteur research units in Bioinformatics and Biostatistics. The list of the services offered by the Hub are detailed hereafter:

  • Short questions : if you have a simple question about bioinformatics and / or statistics, you can ask your question on this website . A mail is sent to all Hub members and one of them will answer you within a few hours.
  • Open desk : come and discuss about your data, your project, your analyses or just about bioinformatics and (bio-)statistics every Tuesday morning from 10 to 12 in Yersin Building (24) 1st Floor. Hub members with complementary fields of expertise will be ready to answer your questions and discuss with you.
  • Collaborative projects : you are starting a project or you are planning experiments? You need advice and / or help in bioinformatics and (bio-)statistics to design your experiment and analyze your data, or develop a specific software? Please submit a project here . Dedicated Hub members can be involved in your project from the beginning.

You may find more information about the Hub of Bioinformatics and Biostatistics on the team webpage .