Challenges

Due to its significance in medicine, many systems biology approaches, such as those for signaling network building, have been developed in recent years. However, there are large limitations for these conventional systems and bioinformatics approaches. Current systems approaches usually extract, accumulate and summarize data from experimental biomedical research literature or integrate existing databases, and connect lists of genes or proteins using these databases. The limitation of the current approaches originates from their assumption that a network is the sum of its components. In systems biology, this assumption does not always hold true. A network is a system in which the components that constitute the network come together, interacting and interdependent, to form a more complex whole unit. A network is usually much greater than the sum of all the components. HarvestTM enables the development of a bioinformatics and biostatistics to provide not only the stored knowledge in systems biology, but also answer questions beyond the stored knowledge that is so urgently needed.

The systems biology approach arises from the need to be able to cope with the size and complexity of biological knowledge and data. This approach enables knowledge to be used within biomedical systems for communication, specification, and other processing tasks. There are limitations for the conventional means of biostatistics and bioinformatics analysis, especially in complex biological systems where various types of information are difficult to integrate. Some of the issues that must be addressed include:

1. The current systems are often like conventional database systems even though the biological knowledge representation is expressive.
2. Little re-use of biological knowledge currently – this is partly because of difficulties in the diversity of their representational form, the explicitness of their semantics, and the range of applications they address.
3. To make expressivity become computability, the knowledge representation of biological models must automatically provide logical inference. Logical inference is what sets knowledge representation apart from a conventional database. Therefore, the biological knowledge with logical inference can answer questions that go beyond what was explicitly told to the system.
4. Development of common biological knowledge patterns and templates can increase the reusability of knowledge.