I agree DSSP is one good starting point. One could look for a region where the solvent accessibility values are all nearly zero. I imagine this would bring a lot of false positives though -- there are often residues that are not quite in the core but are still somewhat buried.

It's not clear whether OP just needs to evaluate this qualitatively or whether an actual algorithm is needed, but I think if we were to turn this into an unsupervised algorithm, we would need to consider the distance from the other buried residues.

Can you please specify what is the difference between a buried residue and a core residue?

Some residues may be buried -- not accessible to solvent -- yet not in the core. Perhaps they could be considered an additional, separate core in this case, but I think the word "core" generally refers to a central continuous region in the protein structure.

I upvoted this question because I think it would be cool to have an app that - given a PDB structure and a solvent accessibility cutoff - could tell you which residues are in the (putative) core. I looked and it doesn't look like this currently exists.

Thanks for the answer. But I think there can be more than one core, as "The definition of a hydrophobic core is a set of residues located within regions of regular secondary structure, whose side chains interact with each other and buried"....Please verify.

Also, it will be really good to know, why do I need to know the solvent accessibility for determining the core?

Also I would really like to thank you for your input :).

You could try chimera: http://www.cgl.ucsf.edu/chimera/download.html

I have read the previous paper but could not work out the algorithm.