Researchers from Europe and the Americas have found certain genes ramp up or wind down in skin, lungs and other body tissues in the hours after death.
They also developed software that can analyse these gene patterns to calculate how long since that person died. Their research was published today in Nature Communications.
What happens to genes after death?
Roderic Guigó, at the Centre for Genomic Regulation in Spain, and an international team decided to investigate how death and the hours afterwards affect gene activity.
Their aim was to create the first comprehensive map of gene activity after death.
The team analysed data from the US-based Genotype-Tissue Expression (GTEx) project. The GTEx repository contains thousands of tissue samples collected from corpses at different time intervals after death, along with their gene activity.
So how do scientists measure gene activity in a human tissue sample?
It’s inferred from levels of mRNA, a type of genetic material similar to DNA. Levels of mRNA increase when a gene’s activity is increased.
Professor Guigó and his crew examined 36 different types of tissue from 540 donors at different intervals after death and looked for gene activity that significantly rose or fell.
They found gene activity in some tissues, such as muscle, dropped off almost immediately after blood stopped flowing.
But in others, not only did gene expression continue, but it ramped up over time.
HBA1 is one gene that ramped up. It encodes a type of haemoglobin, which shuttles oxygen around the body, and consistently increased in 10 tissues tested, including the colon, oesophagus and testis.
After death, when blood stops flowing, cells might ramp up haemoglobin production to scrounge what little oxygen is left, Professor Guigó said.
Developing a genetic stopwatch on death
The researchers then narrowed down the tissue types to the smallest combination that best predicts time of death.
They ended up with four tissue types: sun-exposed skin, fat found just below the skin, thyroid and lung.
To use those tissues in a forensic analysis setting, the team trained a machine-learning algorithm to analyse mRNA data and give the time between death and tissue collection.
When they tested the algorithm, they found its predictions were certainly not perfect, but generally accurate to within an hour. The software is open-sourced.
Forensic scientists currently establish a window of time of death in a range of ways, each of which “aims to narrow that window”, said Brendan Chapman, a forensic biologist at Murdoch University in Perth.
These include taking a body’s deep internal temperature, measuring the stiffness of muscles (rigor mortis), electrically stimulating face muscles to measure excitability, potassium levels in eyeball fluid and taking stock of maggots and other creepy-crawlies.
But, he added, “It’s always been an inexact science”.
“Body temperature, for instance, gives you a window of plus or minus 2.8 hours — at best,” Mr Chapman said.
So the new gene-activity analysis, Mr Chapman said, is “a brilliant foundation” to help investigators calculate time of death, particularly within the first 24 hours.
The major hiccup was sample quality.
“These are clinically collected samples, collected for various other purposes in a clinical environment and well-preserved, ” he said.
And in Australia, that’s a problem.
“Tissue is likely to degrade very quickly, providing a very short window of opportunity to use the gene-expression method to predict [time of death],” Griffith University forensic biologist Kirsty Wright said.
Before the software is added to the forensic toolkit, Professor Guigó said, it does need some additional fine-tuning to take these factors into account.