I´m trying to fit FTIR data (protein secondary structure) with Voigt function (Gaussian-Lorentzian). I was reading about the use of bayesian information criterion (bic) that will penalize adding more components for the fitting but also will take into account the probability of not including the band. I´m looking for an R script to, let´s say, start with some bands and let the algorithm to find a the best fit or to add new bands. Is this possible?