The extent of racemization of aspartic acid (Asp) -expressed as D/L ratio- has been used as a marker of biomolecular degradation in ancient remains. However, Asp racemization rate is highly variable, and depends on biochemical and geochemical factors. In this paper we aim to determine to which extent the fraction analyzed and the kind of sample used may influence the D/L Asp ratios. Other factors, such as burial site and sample preservation conditions, are also considered. D/L Asp ratios were obtained in 38 ancient human samples from 30 different individuals in which Real Time PCR quantification and amplification of short mtDNA fragments had been previously achieved. Four samples were taken from bones, 16 from whole teeth and 18 from dentine. In 7 cases whole tooth and dentine fractions from the same individual were analyzed. The samples belonged to 8 archaeological sites from Pre-pottery Neolithic B (PPNB), Cardial Neolithic and Chalcolithic time periods. Results show significant differences between the D/L Asp ratios obtained in dentine and whole tooth fractions from the same tooth and individual, as well as among dentine samples from the same archaeological site and among samples of the same age from different, though nearby, archaeological sites. Ancient DNA (aDNA) could be characterized in the majority of the samples, independently from their racemization values, which suggests that other factors, apart from the sample preservation stage, are more significantly affecting the racemization rate. A generalized use of the Asp racemization ratio as a threshold value for ancient DNA preservation needs to be questioned until further methodological standardization is considered.