PIPAx
=====
![Widget icon](icons/pipax.png)
Gives access to [**PIPA**](http://pipa.biolab.si/hp/index.html#) databases.
Signals
-------
**Inputs**:
- (None)
**Outputs**:
- **Data**
Selected experiments. Each annotated column contains results
of a single experiment or, if the corresponding option is
chosen, the average of multiple replicates.
Description
-----------
**PIPAx** is a widget for a direct access to [**PIPA**](http://pipa.biolab.si/hp/index.html#) database.
It is very similar to the **GenExpress** and **GEO Data Sets** widgets as it allows you to download the data from
selected experiments.
![PIPA widget](images/PIPAx-stamped.png)
1. Reloads the experiment data.
2. The widget will save (cache) downloaded data, which makes them also available offline. To reset the widget click *Clear cache*.
3. Use *Experiment Sets* to save a selection:
select the experiments, click the "**+**" button and name the
set. To add experiments to the set, click on its name, select
additional experiments and click *Update*.
To remove the set click "**-**".
4. In *Sort output columns* set the attributes by which the output columns are sorted.
Add attributes with a "+" button and remove them
with "-". Switch the sorting order with arrows on the right.
5. Set the expression type for your output data.
- **Raw expression** outputs raw experiment data
- **RPKM expression** outputs data in *reads per kilobase of transcript per million mapped reads*
- **RPKM expression + mapability expression** uses similar normalization, but divides with gene
mapability instead of exon lengths.
The polyA variants use only polyA (mRNA) mapped hits.
6. **Exclude labels with constant values** removes attribute labels that are the same for all selected
experiments from the output data.
**Average replicates (use median)** averages identical experiments by using medians as values.
**Logarithmic (base 2) transformation** computes the log2(value+1) for each value.
7. Click *Commit* to output selected experiments.
8. Log in to access private data.
9. Experiments can be filtered with the *Search* box.
To select which attributes to display right-click on the header. To select multiple experiments
click them while holding the *Control/Command* key.
Example
-------
In the schema below we connected **PIPAx** to **Data Table**, **Set Enrichment**, and **Distance Map**
(through **Distances**) widgets.
![](images/PIPA-Example.png)
The **Data Table** widget above contains the output from the **PIPAx** widget.
Each column contains gene expressions of a single experiment. The labels
are shown in the table header. The **Distance Map** widget shows distances between experiments. The
distances are measured with **Distance** widget, which was set to
compute *Euclidean* distances.