Before attempting to calculate the secondary structure percentages, ascertain whether there are any ligands or cofactors bound to the protein that might contribute to the spectrum.

This is affinity chromatography purified protein. We have only performed dialysis in buffer containing 10mM Tris-HCl pH 7.5 and 10 mM NaF.  

It could be a good starting point to do some literature search, has anyone done CD on this protein previously? Also, if this protein is archived in PDB, then you can find the estimated secondary structure contents or even the simulated CD profile of the protein, and compare the values/plot with yours. Hope this helps.

Hi, I agree with Adam B Shapiro. The spectra does not look like protein, I could not identify the classical maxima and minima for alpha-helix or beta-sheet structures. You can read about it here http://lnbio.cnpem.br/wp-content/uploads/2012/08/CD.pdf

Indeed cofactors or chiral ligands may affect the spectra. If you have a pure protein (as a single band in a silver-stained SDS-PAGE gel), you could try the website for data analysis: http://dichroweb.cryst.bbk.ac.uk/html/home.shtml (you need to register to gain access). 

Another important thing is to follow the voltaje signal associated with ellipticity. Beyond 700-800 mV, the data is unreliable.

Good luck!

Thank you for  your response Adam, Amy and Rogerio. The structure of this protein is available. The CD studies are not performed on this particular protein earlier however homologous protein doesn't show this behavior. The structure has both alpha helices and beta sheets. We were trying to study the temperature stability of this protein in comparison with a homologue. Both proteins are 60 % identical but this one is showing a prominent  dip at 228 nm. The voltage signal data associated with the scan is attached and it is below 500 mV.

There are several reports present which shows the similar behaviour. some all beta proteins ( ex- conconavilin A) shows this type of CD signal. here i attached some article which can solve your problem. some protein aggregate also show the similar pattern