BCA assay. A whole range of reagents and kits available.

agree with BCA method. it measures complexes of BCA with Cu+ ions generated after reaction with peptide bonds, so it is very specific for proteins with no interactions with nucleic acids

Theoredically, mesurements are based on the UV absorbance.

Choose a method that requires longer wave length.

As far as from 260 nm and 280 nm.

UV-visible absorbance is not a good idea in case of crude sample as several other factors absorb below 300nm. On the other hand any of the protein estimation method can solve ur problem.

BCA is measured in 560nm, far from absorbance of nucleic acids

Depends on how accurately you want to know protein concentration. The Lowry, BCA, and Bradford methods are highly idiosyncratic with regard to amino acid content of proteins, but are fine for crude estimations of protein concentration. Any reducing agents present will interfere with Lowry or BCA, detergents and lipids will interfere with Bradford. UV methods are superior for purified proteins (there are several), but these are probably not relevant for you if you have a highly heterogeneous mixture. The microbiuret method is the most accurate of the colorimetric methods in my opinion, but there are also some interferences from thiols, and it is not the most sensitive method.

Alternatively,try Pierce 660 nm Protein assay. In contrast to BCA, this is largely unaffected by most buffer components and detergents.(

http://www.piercenet.com/browse.cfm?fldID=C16B4598-5056-8A76-4E7D-ADABD3720DFE)

Go for TCA precipitation first. It will isolate proteins from any mixture, though may denature your proteins. For quantification, it should not be a matter. Afterward, make dilution as same as before using ddH2O, followed by BCA assay.