<p>This study aimed to develop an accurate and practical DNA methylation (DNAm) frequency-based chronological age estimation model for bottlenose dolphins off the Pacific coast of Japan. The skin tissues of 28 individuals with known chronological age determined by their teeth were provided for DNA extraction. The DNAm analysis was performed using a next-generation sequencer for a total of 24 CpG sites in three genes (<i>TET2</i>, <i>GRIA2</i>, and <i>CDKN2A</i>) that are known to be chronological-age-related. The best age estimation model utilizing the DNAm frequencies (%) was selected as follows: estimated chronological age = 7.7086 + 3.2756 × <i>GRIA2</i>_CpG4&#xa0;−&#xa0;2.2931 × <i>CDKN2A</i>_CpG1 (<i>R</i><sup>2</sup> = 0.86, and a mean absolute error of 2.74&#xa0;years from the age assessed by the teeth). Two age estimation models from the different populations (this study and one previously reported in Sarasota Bay, FL, USA) were compared by exchanging the respective samples for the age estimation. In both cases, the accuracy of age estimation decreased when the sample was applied to the non-original model. This was because the CpG sites selected for model development differed between populations. Thus, this study indicates that DNAm-based age estimation for a population is more accurate when conducted in a population-specific manner.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Inter-population variability in DNA methylation-based chronological age estimation model described from common bottlenose dolphin Tursiops truncatus

  • Aoi Irie,
  • Hikari Maeda,
  • Atsushi Tanabe,
  • Hiroeki Sahara,
  • Reiko Nozaki,
  • Hidehiro Kondo,
  • Yu Kanaji,
  • Naohisa Kanda,
  • Hiroto Murase

摘要

This study aimed to develop an accurate and practical DNA methylation (DNAm) frequency-based chronological age estimation model for bottlenose dolphins off the Pacific coast of Japan. The skin tissues of 28 individuals with known chronological age determined by their teeth were provided for DNA extraction. The DNAm analysis was performed using a next-generation sequencer for a total of 24 CpG sites in three genes (TET2, GRIA2, and CDKN2A) that are known to be chronological-age-related. The best age estimation model utilizing the DNAm frequencies (%) was selected as follows: estimated chronological age = 7.7086 + 3.2756 × GRIA2_CpG4 − 2.2931 × CDKN2A_CpG1 (R2 = 0.86, and a mean absolute error of 2.74 years from the age assessed by the teeth). Two age estimation models from the different populations (this study and one previously reported in Sarasota Bay, FL, USA) were compared by exchanging the respective samples for the age estimation. In both cases, the accuracy of age estimation decreased when the sample was applied to the non-original model. This was because the CpG sites selected for model development differed between populations. Thus, this study indicates that DNAm-based age estimation for a population is more accurate when conducted in a population-specific manner.