Quality Control
===============

![Quality Control widget icon](icons/quality-control.png)

Computes and plots distances between experiments or replicates.

Signals
-------

**Inputs**:

- **Data**

  Data set.

**Outputs**:

- (None)

Description
-----------

**Quality Control** measures distances between experiments (usually replicates) for a selected label. The widget visualizes
distances by selected label. Experiments that lie the farthest from the initial black line
should be inspected for anomalies.

![image](images/QualityControl-stamped.png)

1. Information on the input.
2. Separate experiments by label.
3. Sort experiments by label.
4. Compute distances by:
   - [**Pearson correlation**](https://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient)
   - [**Euclidean distances**](https://en.wikipedia.org/wiki/Euclidean_distance)
   - [**Spearman correlation**](https://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient)
5. Hover over the vertical line to display the information on a chosen instance. Click on the black line to
   change the reference to that instance.

Example
-------

**Quality Control** widget gives us feedback on the quality of our data and can be connected to any widget with
data output. In the example above (see the image under *Description*) we fed 9 experiments of 
*Cyclic AMP pulsing* of *Dictyostelium discoideum* from **GenExpress** widget into **Quality Control**
and separated them by timepoint label. We found replicate 2 from tp 2 among the tp 1 data, meaning
we should inspect these data further.