CausalTrail is a tool for causal hypotheses testing using causal bayesian networks (CBN) and the do-calculus. CausalTrail provides a console application and a graphical user interface.

Stöckel D, Schmidt F, Trampert P and Lenhof HP. CausalTrail: Testing hypothesis using causal Bayesian networks [version 1; referees: awaiting peer review] F1000Research 2015, 4(ISCB Comm J):1520 doi: 10.12688/f1000research.7647.1

The current version of the paper is available here.

CausalTrail can be build using cmake.

###Mandatory Dependencies### The Boost library as well as a C++ compiler supporting C++14 have to be available to build the console version of CausalTrail.

Supported compilers are:

• >= GCC 4.9
• >= Clang 3.5

###Optional Dependencies### To build CausalTrails unit test suite, gtest >= 1.7.0 is required. To build the GUI, Qt version 5.4 or higher has be installed.

###Step By Step Build Commands### Enter CausalTrails directory and execute the following:

mkdir build
cd build
cmake ..

If the path to gtest can not be found automatically, specify it via

cmake . -DGTEST_SRC_DIR=<path>

If the path to Qt can not be found automatically, specify it via

cmake . -DQt5Widgets_DIR=<path>

Build the project by typing

make

To use multiple cores (e.g. 4) for building use the -j option:

make -j 4

The executable file for the console version is located in the folder build/core, the GUI version is located in the folder build/gui, and the tests can be found in the folder build/test.

To run CausalTrails tests type

make test

The console version of CausalTrail can be evoked with the command

./CausalTrail <Observations.txt> <Discretisation_Information.json> <Network.tgf>

or alternatively

./CausalTrail <Observations.txt> <Discretisation_Information.json> <Network.sif> <Network.na>

We provide details on the input files in the next section.

The GUI can be launched with

./CausalTrailGui

###Network Files### We support two kinds of network formats: the trivial graph format (tgf) and the simple interaction format (sif) along with node node atrribute (na) files.

####Trivial Graph Format####

The tgf format has the following structure:

NodeID NodeAttribute

...

#

NodeID NodeID EdgeAtribute

The upper part of a tgf file contains the mapping between node identifiers and at most one optional attribute, e.g. node names. The # marks the beginning of the actual network definition. Edges are directed from the first to the second node identifier. An edge between two nodes can be mapped to at most one optional attribute.

#####Example#####

1 Apples
2 Bananas
3 Pears
#
1 2
2 3
3 1

####Node Attribute Format#### Node attribute files have the following structure:

AttributeName (class = Type)
NodeID =	NodeAttribute
...

Different node attribute classes can be referenced via the AttributeName. Type states the data type of the node attributes. The mapping of NodeID to NodeAttribute has to be unique within one class of attributes.

#####Example##### NodeName (class=java.lang.String) 1 = Apples 2 = Bananas 3 = Pears

####Simple Interaction Format#### The simple interaction format is structured as follows:

...
NodeID EdgeType NodeID
...

Nodes in the network are identified via the NodeID. Therefore, the NodeIDs have to be unique. The left NodeID represents the source of an edge, the right on represents the target. It is possible to assign more than one target node to a distinct source node, so multiple edges can be stored in one line. The EdgeType encodes the type of an edge, e.g. whether an edge between two nodes is directed or not. It is also common to encode biological meaning in the EdgeType. For example, pd represents Protein-DNA interactions, whereas pp represents Protein-Protein interactions. The EdgeType can also be a longer string, allowing the encoding of more complex descriptions, e.g. activates, inactivates, or phosphorylates. If it is not necessary to encode any specific meaning for an edge xx or yy can be used as an EdgeType.

#####Example##### 1 pp 2 2 pp 3 3 pp 1

For more information on the SIF format see here.

###Data Files### The samples from which we learn the causal Bayesian network parameters should be provided in a tab-delimited text file where the columns represent the samples and the rows represent the features. An example is shown below.

SNV1		Yes	No	Yes	No
SNV2		No 	No	No	Yes
Expression	1.7	1.2	1.4	0.6

###Discretisation Information### CausalTrail uses discretised input data for training the node parameters. As measurements often come as continuous values, they have to be discretised before the actual parameter learning can take place. For this reason, we provide several discretisation methods within CausalTrail.

Using a simple json file, the user can control the discretisation process. For example:

{
	"SNV1":
	{
		"method":"None",
	},
	"SNV2":
	{
		"method":"None",
	},
	"Expression":
	{
		"method":"Threshold",
		"threshold": "1.0"
	}
}

For every node in the network, the method field specifies the discretisation method. If necessary, additional fields can be used to provide specific information that is needed for the discretisation, e.g. a manually determined threshold as shown in the example above. A list of all keywords and discretisation methods is shown below.

• Floor: Round the data to the next smaller integer value.
• Ceil: Round the data to the next larger integer value.
• Round: Round the data.
• ArithmeticMean: Divide the data into two groups according to the arithmetic mean.
• HarmonicMean: Divide the data into two groups according to the harmonic mean.
• Median: Divide the data into two groups according to the median.
• Threshold:
  • Divide the data into two groups according to a manually defined threshold.
  • The actual threshold is specified in the field threshold.
• BracketMedians:
  • Discretise the data according to the bracket medians method.
  • The number of buckets is specified in the field buckets.
• PearsonTukey: Discretise the data according to the Pearson Tukey method.
• Z-Score: Divide the data into two groups according to the Z-Score.
• None: The data is already discrete.

The initial layout of CausalTrails GUI is shown below. At the bottom of the window, there is a dock widget containing general information on the current session, labelled Log. As we see later, the middle area is used for network visualisation and query management. At the top, there is a toolbar allowing direct access to the most important actions. Buttons are enabled according to the current status of a session. Thus, errors caused by wrong user input can be avoided. In addition to the toolbar, there is a menubar allowing access to all functions of CausalTrail.

In the following we provide step by step guidelines on how to use our tool. For illustration purposes, we use the Student Network presented in Probabilistc Graphical Models by Koller and Friedman.

###Network Management### Networks can be loaded by a click on Load Network in the toolbar or by clicking on Network -> Load Network in the menu. A dialog will be shown to select networks represented in the formats introduced above. The dialog can also be opened by pressing Strg + O.

Upon loading, the network layout is computed using graphviz. If graphviz is not available, the layout is generated by a force directed layouting algorithm included in CausalTrail.

The network view is interactive, e.g. it is possible to move nodes or to zoom in or out the network visualisation. By clicking on Layout in the toolbar, the menubar or by pressing Strg + L, the network is layouted again.

Networks are visualised in a tab window. In case that the user loads multiple networks, each network is shown in its own tab. Using CausalTrails svg export function, a network visualisation can be exported to a svg file. To do so, click Network -> Export SVG.

A click on Delete Network deletes the network that is currently shown. This can also be done with Strg + D.

An example of the visualisation of the Student Network is depicted below.

To load the Student Network, use the files:

test/data/Student.na
test/data/Student.sif

###Loading Samples and Training the Network To load samples, click on Load Samples in the toolbar or in the menu. Once a suitable file is chosen, the data is shown in a table allowing manual inspection of the data as well as (de)selection of individual samples. This allows the exclusion of distinct samples from the analysis. An example for the student network is shown below.

Upon confirming the data by a click on OK, a window for selection of discretisation methods is shown.

Here, the user has two options: Either the discretisation information is loaded from a existing json file, or it is specified using the interface. A json file can be loaded by clicking on Load. In order to simplify the discretisation step, manually specified discretisation information can be stored in a json file by clicking on Save. As soon as the discretisation information is specified, the user can continue with parameter learning by clicking OK.

During parameter learning the conditional probability tables (CPT) for all nodes are computed. It is possible to look at the individual CPTs of each node by right clicking on a node and selecting Show CPT in the popup-menu.

Once the learning is completed, two new dock widgets are shown, the Query History and the Query Control Panel.

The Query History lists all submitted and valid queries for individual networks. The Query Control Panel is used in formulating queries. We provide several examples for query management in the next section.

To train the Student Network use the files:

test/data/StudentData.txt
test/data/controlStudent.json

###Query Management### ####Overview#### CausalTrail offers four ways to submit a query:

1. A query can be entered directly, as in the console version of CausalTrail.
2. A query can be entered interactively.
3. A query can be chosen from the Query History.
4. Several queries can be executed using a Query Batch File.

####Direct Query Formulation#### Queries can be entered directly into the Edit field at the top of the Query Control Panel. Correctness of queries is checked while typing. The background colour of the Edit field switches to green if the query is correct and to red otherwise. Queries can be submitted by a click on the green tick next to the Edit field or by pressing Enter.

Entering queries directly requires the user to be familiar with our query language in detail. As this can not be expected from the general user, we offer an interactive query construction introduced in the next section.

####Interactive Query Formulation#### To facilitate the formulation of queries, CausalTrail supports interactive query construction. To build a query, the user has to move the mouse over a node of interest and perform a right click on it. A context menu allowing the following operations is shown:

1. Probability of: Calculates the probability that this node obtains the specified value.
2. Arg Max: Calculates the most likely value assignment for this node. Note that it is not possible to combine (1) and (2) in one query.
3. Condition on: Calculates the conditional probability of other nodes given the specified value of the currently selected node.
4. Set value to: Perform a do-intervention on the current node. The edges to the nodes parents are deleted and its own value is fixed to the selected value.
5. Add Edge to: Adds an edge to another node. To add a new edge double click on the desired target node. The added edge will appear in red. If the new edge would induce a cycle, an error will be shown in the Log. Adding an edge causes retraining of the entire network.

Once an item is selected, it is shown in one of the boxes in the Query Control Panel. A colour code and a natural language wrapper around the query item boxes help to understand the query. Double clicking on an item in the Query Control Panel removes it from the current query.

In addition to the operations on nodes, there is an operation on edges. A right click on an edge opens a context menu allowing to remove the selected edge. Removed edges are shown in grey. As for adding an edge, removing one causes retraining of the network too.

####Query History#### The query history enables the user to reload a previously submitted query. There are two ways to do this:

1. A query can be directly selected via a double click in the Query History.
2. Using the Back and Forward arrows next to the query Edit field one can go stepwise forward or backward in query history.

####Query Batch Files#### To permit the user to quickly process a set of queries on networks trained on different data sets, CausalTrails offers Query Batch Files. A batch file containing all queries currently shown in the Query History can be created by clicking on Create Batchfile. It can be executed by a click on Execute Batchfile. Results are shown in the Log.

####Examples#### To illustrate the usage of CausalTrail further, we present a few example queries in the Student Network.

#####Predictions###### In this example, we compute the probability that Intelligence obtains the value i1.

In the second example, we compute the probability that Intelligence obtains the value i1 if Grade has value g1 and SAT has value s1.

#####Interventions##### Here, we compute the probability that Intelligence obtains the value i1, if we perform a do-intervention on Grade, setting its value to g1 and given that SAT has value s1.

#####Counterfactuals##### In the last example, we compute the probability to get a letter, if we have not received a letter before.

###Sessions### To avoid repeating the process of network and sample loading, CausalTrail supports sessions. A session in CausalTrail contains all currently trained networks and submitted queries. To save a session, click on Save Session in the toolbar or click File -> Save Session in the menu. A session can be restored by a click on Load Session in the toolbar or by clicking File -> Load Session in the menu.