Investigating food production‐associated DNA methylation changes in paleogenomes: Lack of consistent signals beyond technical noise

• Published in: Evolutionary applications Vol. 17; no. 7; pp. e13743 - n/a
• Main Authors: Çokoğlu, Sevim Seda, Koptekin, Dilek, Fidan, Fatma Rabia, Somel, Mehmet
• Format: Journal Article
• Language: English
• Published: England John Wiley & Sons, Inc 01.07.2024; John Wiley and Sons Inc; Wiley
• Subjects: ancient DNA; CpG islands; Diet; DNA damage; DNA methylation; Epigenetics; Food; Genomes; genomics/proteomics; human evolution; Laboratories; Lifestyles; Neolithic; Neolithic transition; Noise; Nutrient deficiency; Original; Physiology; Population density; Population genetics; Software; Eurasia; DNA methylation; ancient DNA; genomics/proteomics; human evolution; Neolithic transition; epigenetics
• DOI: 10.1111/eva.13743
• ISSN: 1752-4571

The Neolithic transition introduced major diet and lifestyle changes to human populations across continents. Beyond well‐documented bioarcheological and genetic effects, whether these changes also had molecular‐level epigenetic repercussions in past human populations has been an open question. In fact, methylation signatures can be inferred from UDG‐treated ancient DNA through postmortem damage patterns, but with low signal‐to‐noise ratios; it is thus unclear whether published paleogenomes would provide the necessary resolution to discover systematic effects of lifestyle and diet shifts. To address this we compiled UDG‐treated shotgun genomes of 13 pre‐Neolithic hunter‐gatherers (HGs) and 21 Neolithic farmers (NFs) individuals from West and North Eurasia, published by six different laboratories and with coverage c.1×–58× (median = 9×). We used epiPALEOMIX and a Monte Carlo normalization scheme to estimate methylation levels per genome. Our paleomethylome dataset showed expected genome‐wide methylation patterns such as CpG island hypomethylation. However, analyzing the data using various approaches did not yield any systematic signals for subsistence type, genetic sex, or tissue effects. Comparing the HG‐NF methylation differences in our dataset with methylation differences between hunter‐gatherers versus farmers in modern‐day Central Africa also did not yield consistent results. Meanwhile, paleomethylome profiles did cluster strongly by their laboratories of origin. Using larger data volumes, minimizing technical noise and/or using alternative protocols may be necessary for capturing subtle environment‐related biological signals from paleomethylomes.