RT Journal
A1 Fernandes, Daniel M.
A1 Sirak, Kendra A.
A1 Cheronet, Olivia
A1 Novak, Mario
A1 Brück, Florian
A1 Zelger, Evelyn
A1 Llanos-Lizcano, Alejandro
A1 Wagner, Anna
A1 Zettl, Anna
A1 Mandl, Kirsten
A1 Duffet Carlson, Kellie Sara
A1 Oberreiter, Victoria
A1 Özdoğan, Kadir T.
A1 Sawyer, Susanna
A1 La Pastina, Francesco
A1 Borgia, Emanuela
A1 Coppa, Alfredo
A1 Dobeš, Miroslav
A1 Velemínský, Petr
A1 Reich, David
A1 Bell, Lynne S.
A1 Pinhasi, Ron
T1 Density separation of petrous bone powders for optimized ancient DNA yields
JF Genome Research 
JO Genome Research 
YR 2023 
FD April 01 
VO 33 
IS 4 
SP 622 
OP 631 
DO 10.1101/gr.277714.123 
UL http://genome.cshlp.org/content/33/4/622.abstract 
AB Density separation is a process routinely used to segregate minerals, organic matter, and even microplastics, from soils and sediments. Here we apply density separation to archaeological bone powders before DNA extraction to increase endogenous DNA recovery relative to a standard control extraction of the same powders. Using nontoxic heavy liquid solutions, we separated powders from the petrous bones of 10 individuals of similar archaeological preservation into eight density intervals (2.15 to 2.45 g/cm3, in 0.05 increments). We found that the 2.30 to 2.35 g/cm3 and 2.35 to 2.40 g/cm3 intervals yielded up to 5.28-fold more endogenous unique DNA than the corresponding standard extraction (and up to 8.53-fold before duplicate read removal), while maintaining signals of ancient DNA authenticity and not reducing library complexity. Although small 0.05 g/cm3 intervals may maximally optimize yields, a single separation to remove materials with a density above 2.40 g/cm3 yielded up to 2.57-fold more endogenous DNA on average, which enables the simultaneous separation of samples that vary in preservation or in the type of material analyzed. While requiring no new ancient DNA laboratory equipment and fewer than 30 min of extra laboratory work, the implementation of density separation before DNA extraction can substantially boost endogenous DNA yields without decreasing library complexity. Although subsequent studies are required, we present theoretical and practical foundations that may prove useful when applied to other ancient DNA substrates such as teeth, other bones, and sediments.