Today it is possible to analyze hundreds of samples simultaneously, representing a variety of biological conditions. These large-scale approaches raise new challenges regarding experimental design. In particular, it is becoming necessary to overcome "batch effects" and to limit certain biases through "randomization" methods.

The distribution of samples, prior to their biochemical preparation, is one of the many factors that can bias results in clinical research. To simplify this tedious task, researchers at Irig have developed an easy-to-use application named Well Plate Maker.

This approach differs from existing methods – which locate samples by their coordinates on a plate – in the following ways:

• it includes a graphical interface that facilitates its use without requiring any programming skills;
• and it can be used upstream of the experiment to optimize sample distribution and to better design the well plates.

 For this, additional information must be provided and the following defined:

•  criteria for belonging to different groups;
• spatial or neighboring constraints;
• and the size and number of plates.
  

In the case of multiple plates, the samples are uniformly distributed among the different plates according to the total sample number and group size, making it possible to avoid "plate effects". Furthermore, Well Plate Maker simplifies block randomization in order to limit other types of batch effects.

Researchers at Irig use the application to optimally distribute plasma samples from clinical cohorts prior to their biochemical preparation and proteomic analysis by mass spectrometry.

 With its extensive array of options, Well Plate Maker can be adapted to a wide range of applications (various omics analyses, screenings, etc.).

This work was conducted with the support of the ANR.