When used correctly, they can help us unpick some of the mysteries of our biology, and our mortality. Be honest: Have you ever looked up someone from your childhood on social media with the sole intention of seeing how they've aged? One of my colleagues, who shall remain nameless, certainly has. He recently shared a photo of a former classmate. "Can you believe we're the same age?" he asked, with a hint of glee in his voice. A relative also delights in this pastime. "Wow, she looks like an old woman," she'll say when looking at a picture of someone she has known since childhood. The years certainly are kinder to some of us than others. But wrinkles and gray hairs aside, it can be difficult to know how well--or poorly--someone's body is truly aging, under the hood. A person who develops age-related diseases earlier in life, or has other biological changes associated with aging (such as elevated cholesterol or markers of inflammation), might be considered "biologically older" than a similar-age person who doesn't have those changes. Some 80-year-olds will be weak and frail, while others are fit and active. Longevity clinics offer a mix of services that largely cater to the wealthy.

Interrogating the evolution of biological changes at early stages of life requires longitudinal profiling of molecules, such as DNA methylation, which can be challenging with children. We introduce a probabilistic and longitudinal machine learning framework based on multi-mean Gaussian processes (GPs), accounting for individual and gene correlations across time. This method provides future predictions of DNA methylation status at different individual ages while accounting for uncertainty. Our model is trained on a birth cohort of children with methylation profiled at ages 0-4, and we demonstrated that the status of methylation sites for each child can be accurately predicted at ages 5-7. We show that methylation profiles predicted by multi-mean GPs can be used to estimate other phenotypes, such as epigenetic age, and enable comparison to other health measures of interest. This approach encourages epigenetic studies to move towards longitudinal design for investigating epigenetic changes during development, ageing and disease progression.