Proteins play a major role in many biological processes. Understanding their functional properties is key in order to develop an ability to manipulate their properties for beneficial use. Recent advances in high throughput technology has led to the vast quantities of biological data becoming available. This in turn has to the development of sophisticated computational tools to analyse and process this data of high volume and complexity. The computational tools have been used both in a complimentary manner to the wet lab based experiments and as independent research utilities.Most of the work within this thesis focuses on using computational methods to expand the limits of our knowledge of functional and structural properties of different families of proteins. First part of the thesis deals with classifying proteins that are dependent on NAD and NADP for their function. A correlation between the conformation of the cofactor within the active site and the protein function was discovered. Second part of the thesis focuses on studying pH dependence in acid sensing ion channels using comparative modelling and continuum electrostatics based computational methods. The key pH sensing residues were identified within a complex mechanism of channel operation. The variation in pH dependent properties of different ASIC family members and other related ion channels was discovered to be related to the number of acid sensors present. In the last part of the thesis, computational methods are used to assess the likelihood of presence of the radical character in the catalytic mechanism of flavin utilising enzymes.