Boolean networks represent a popular formalism for description of regulatory networks modelling dynamical processes occuring in complex biological systems. One of the key problems is to reconstruct such models from various kinds of experimental data (e.g., time-series, steady-state). Recently, there have been developed several methods for automatised inference of Boolean networks. Most of these methods target synchronous semantics of Boolean networks. However, synchronous models are known to be not sufficient to capture the dynamics of biological systems.

The goal of this thesis is to extend the inference problem to the settings that is not primarily restricted to synchronous semantics. In particular, based on specific types data it might be necessary to consider behaviour that cannot be achieved with synchronous semantics. The general goal is to design a framework for inference of BNs that will reflect the particular semantics based on the input data.

Concretely, the student will target the following objectives:

• to review existing methods for inference of BNs from data,
• to identify the limits in existing methods in relation to the specific types of input data,
• to propose extensions/improvements of existing algorithms that will (partially) overcome some of the identified limits.

The student is encouraged to make a prototype implementation of key parts of the proposed extension.