Translating time from transcriptomics across the lifespan in humans, chimpanzees and macaques
1. Department of Computer Science, University of Bath, UK.
2. Center for Neuroscience, Delaware State University, Dover, DE, USA
3. Department of Biology, Bath Spa University, Bath Spa, United Kingdom
4. College of Veterinary Medicine, Auburn University, USA.
There are still unanswered questions as to how the human brain evolved. To investigate the evolution of human cognition, it is crucial to align equivalent ages across primates. Therefore, appropriate procedures to model human and non-human primate ages are required. The aim of the larger study is to include time points across a range of scales including transcriptomics, behaviour, and anatomy to investigate biological pathways in primates. In this analysis, the differential gene expression in brain tissue from the human, chimpanzee, and macaque samples, across the lifespan of each species, was examined. The bulk RNA- sequencing data used in this study encompassed, 78 human, 74 chimpanzee, and 31 macaque samples with 11622 gene rows. The age range for the human, macaque, and chimpanzee species samples was 0-76, 0-46, 0-21 years respectively.
Machine learning models were implemented to predict age from gene expression. The following machine learning models were applied to the combined dataset: Elastic net (glmnet), support vector machine (SVM) and Gaussian process regression (GPR) and Random Forest (RF). In terms of prediction accuracy, the glmnet, SVM and GPR models were all found to outperform the RF model. However, RF and glmnet were the most interpretable for deducing the most important genes for predicting age from the gene expression patterns. The variation in transcription revealed that old age is unusually extended in humans and therefore further evaluation is necessary to understand ageing in humans versus non-human primates.