The tumor necrosis factor (TNF) is a complex protein that plays a very important role in a number of biological functions including apoptotic cell death, tumor regression, cachexia, inflammation inhibition of tumorigenesis and viral replication. Its most interesting function is that it is an inhibitor of tumorigenesis and inductor of apoptosis. Thus, the TNF could be a good candidate for cancer therapy. However, the TNF has also inflammatory and toxic effects. Therefore, it would be very important to understand complex functions of the TNF and consequently be able to predict mutations or even design the new TNF-related proteins that will have only a tumor inhibition function, but not other side effects. This can be achieved by applying the resonant recognition model (RRM), a unique computational model of analysing macromolecular sequences of proteins, DNA and RNA. The RRM is based on finding that certain periodicities in distribution of free electron energies along protein, DNA and RNA are strongly correlated to the biological function of these macromolecules. Thus, based on these findings, the RRM has capabilities of protein function identification, prediction of bioactive amino acids and protein design with desired biological function. Using the RRM, we separate different functions of TNF as different periodicities (frequencies) within the distribution of free energy electrons along TNF protein. Interestingly, these characteristic TNF frequencies are related to previously identified characteristics of proto-oncogene and oncogene proteins describing TNF involvement in oncogenesis. Consequently, we identify the key amino acids related to the crucial TNF function, i.e. receptor recognition. We have also designed the peptide which will have the ability to recognise the receptor without side effects.
Keywords: Amino acid; DNA; Fast Fourier transform; Oncogenes; Protein; Protein modelling; RNA; Resonant recognition model; Tumor necrosis factor.