We exploit the availability of recent experimental data on a variety of proteins to develop a Web-based prediction algorithm (BPPred) to calculate several biophysical parameters commonly used to describe the folding process. These parameters include the equilibrium m-values, the length of proteins, and the changes upon unfolding in the solvent-accessible surface area, in the heat capacity, and in the radius of gyration. We also show that the knowledge of any one of these quantities allows an estimate of the others to be obtained, and describe the confidence limits with which these estimations can be made. Furthermore, we discuss how the kinetic m-values, or the Beta Tanford values, may provide an estimate of the solvent-accessible surface area and the radius of gyration of the transition state for protein folding. Taken together, these results suggest that BPPred should represent a valuable tool for interpreting experimental measurements, as well as the results of molecular dynamics simulations.Keywords: protein denaturation; urea; guanidine hydrochloride; guanidinium chloride; protein folding; m-values; SASA; radius of gyration; heat capacity; transition state; unfolded state; denatured state The possibility of interpreting quantities readily measurable experimentally in terms of descriptors of protein strucure has contributed very significantly to our understanding of the folding process. In a seminal work, Myers et al. (1995) considered an earlier suggestion by Schellman (1978) and showed that the change in solvent-accessible surface area (DSASA) upon unfolding is related linearly to the experimental m D-N -value (Pace 1986), which describes how the stability nG D-N of the native state of a protein decreases linearly with the concentration of denaturant (Tanford 1968(Tanford , 1970:The relationship between m-values and DSASA is extremely useful because it gives important insights into the determinants of protein stability and the equilibrium properties of proteins. In order to establish such a relationship, however, one needs an estimate of the value of SASA of the unfolded state. Despite recent advances (Mok et al. 2005), it is still very challenging to measure SASA directly in the denatured state. Myers et al. (1995) derived it from a tripeptide model (Shrake and Rupley 1973;