Kristian Hanghøj1,2, Gabriel Renaud1, Anders Albrechtsen3, Ludovic Orlando1,2. 1. Lundbeck Foundation GeoGenetics Center, University of Copenhagen, Øster Voldgade 5-7, 1350K Copenhagen, Denmark. 2. Laboratoire d'Anthropobiologie Moléculaire et d'Imagerie de Synthèse, CNRS UMR 5288, Université de Toulouse III, Paul Sabatier (UPS), 31000 Toulouse, France. 3. Computational and RNA Biology, Department of Biology, University of Copenhagen, 2200 Copenhagen, Denmark, Øster voldgade 5-7, 1350k.
Abstract
BACKGROUND: Recent computational advances in ancient DNA research have opened access to the detection of ancient DNA methylation footprints at the genome-wide scale. The most commonly used approach infers the methylation state of a given genomic region on the basis of the amount of nucleotide mis-incorporations observed at CpG dinucleotide sites. However, this approach overlooks a number of confounding factors, including the presence of sequencing errors and true variants. The scale and distribution of the inferred methylation measurements are also variable across samples, precluding direct comparisons. FINDINGS: Here, we present DamMet, an open-source software program retrieving maximum likelihood estimates of regional CpG methylation levels from ancient DNA sequencing data. It builds on a novel statistical model of post-mortem DNA damage for dinucleotides, accounting for sequencing errors, genotypes, and differential post-mortem cytosine deamination rates at both methylated and unmethylated sites. To validate DamMet, we extended gargammel, a sequence simulator for ancient DNA data, by introducing methylation-dependent features of post-mortem DNA decay. This new simulator provides direct validation of DamMet predictions. Additionally, the methylation levels inferred by DamMet were found to be correlated to those inferred by epiPALEOMIX and both on par and directly comparable to those measured from whole-genome bisulphite sequencing experiments of fresh tissues. CONCLUSIONS: DamMet provides genuine estimates for local DNA methylation levels in ancient individual genomes. The returned estimates are directly cross-sample comparable, and the software is available as an open-source C++ program hosted at https://gitlab.com/KHanghoj/DamMet along with a manual and tutorial.
BACKGROUND: Recent computational advances in ancient DNA research have opened access to the detection of ancient DNA methylation footprints at the genome-wide scale. The most commonly used approach infers the methylation state of a given genomic region on the basis of the amount of nucleotide mis-incorporations observed at CpG dinucleotide sites. However, this approach overlooks a number of confounding factors, including the presence of sequencing errors and true variants. The scale and distribution of the inferred methylation measurements are also variable across samples, precluding direct comparisons. FINDINGS: Here, we present DamMet, an open-source software program retrieving maximum likelihood estimates of regional CpG methylation levels from ancient DNA sequencing data. It builds on a novel statistical model of post-mortem DNA damage for dinucleotides, accounting for sequencing errors, genotypes, and differential post-mortem cytosine deamination rates at both methylated and unmethylated sites. To validate DamMet, we extended gargammel, a sequence simulator for ancient DNA data, by introducing methylation-dependent features of post-mortem DNA decay. This new simulator provides direct validation of DamMet predictions. Additionally, the methylation levels inferred by DamMet were found to be correlated to those inferred by epiPALEOMIX and both on par and directly comparable to those measured from whole-genome bisulphite sequencing experiments of fresh tissues. CONCLUSIONS: DamMet provides genuine estimates for local DNA methylation levels in ancient individual genomes. The returned estimates are directly cross-sample comparable, and the software is available as an open-source C++ program hosted at https://gitlab.com/KHanghoj/DamMet along with a manual and tutorial.
Authors: Adrian W Briggs; Udo Stenzel; Matthias Meyer; Johannes Krause; Martin Kircher; Svante Pääbo Journal: Nucleic Acids Res Date: 2009-12-22 Impact factor: 16.971
Authors: Matthias Meyer; Martin Kircher; Marie-Theres Gansauge; Heng Li; Fernando Racimo; Swapan Mallick; Joshua G Schraiber; Flora Jay; Kay Prüfer; Cesare de Filippo; Peter H Sudmant; Can Alkan; Qiaomei Fu; Ron Do; Nadin Rohland; Arti Tandon; Michael Siebauer; Richard E Green; Katarzyna Bryc; Adrian W Briggs; Udo Stenzel; Jesse Dabney; Jay Shendure; Jacob Kitzman; Michael F Hammer; Michael V Shunkov; Anatoli P Derevianko; Nick Patterson; Aida M Andrés; Evan E Eichler; Montgomery Slatkin; David Reich; Janet Kelso; Svante Pääbo Journal: Science Date: 2012-08-30 Impact factor: 47.728
Authors: Qiaomei Fu; Heng Li; Priya Moorjani; Flora Jay; Sergey M Slepchenko; Aleksei A Bondarev; Philip L F Johnson; Ayinuer Aximu-Petri; Kay Prüfer; Cesare de Filippo; Matthias Meyer; Nicolas Zwyns; Domingo C Salazar-García; Yaroslav V Kuzmin; Susan G Keates; Pavel A Kosintsev; Dmitry I Razhev; Michael P Richards; Nikolai V Peristov; Michael Lachmann; Katerina Douka; Thomas F G Higham; Montgomery Slatkin; Jean-Jacques Hublin; David Reich; Janet Kelso; T Bence Viola; Svante Pääbo Journal: Nature Date: 2014-10-23 Impact factor: 49.962
Authors: Jakob Skou Pedersen; Eivind Valen; Amhed M Vargas Velazquez; Brian J Parker; Morten Rasmussen; Stinus Lindgreen; Berit Lilje; Desmond J Tobin; Theresa K Kelly; Søren Vang; Robin Andersson; Peter A Jones; Cindi A Hoover; Alexei Tikhonov; Egor Prokhortchouk; Edward M Rubin; Albin Sandelin; M Thomas P Gilbert; Anders Krogh; Eske Willerslev; Ludovic Orlando Journal: Genome Res Date: 2013-12-03 Impact factor: 9.043