The biomarkers in this study, he said, could provide a more objective and accurate measure of food intake, allowing scientists to pin-point any correlation with pregnancy complications and the health of mother and baby.
Samples of hair and blood from 1200 Singaporean women have already begun arriving at the centre's laboratories.
Gravida researchers had already established how biomarker profiles from hair in early pregnancy could predict which women were going to go on and suffer complications - "we are now using our advantages," Professor Baker said.
"Once concluded, the best outcome from our study is that we may be able to say: if you eat this during pregnancy, your metabolic profile will look like this, and you will have a higher risk of getting gestational diabetes, or having an early birth, or having one of a number of other complications in pregnancy."
The causes of such conditions - among them life-threatening pre-eclampsia, which affects up to 5 per cent of all pregnancies - were challenging to address given the number of potential factors at play.
"Pre-eclampsia, certainly, is a disease with many different pathways, but it would be very surprising if dietary and environmental influences didn't contribute to conditions like it."
Professor Baker didn't think it far-fetched to propose food could influence the health of a pregnancy - indeed, scientists were increasingly finding food played a major part.
The question was, he said, which food for what mother?
"It doesn't take a rocket scientist to realise what you are really dealing with is that different diets suit different people in different populations and at different times of their life."
An ideal outcome would allow clinicians to identify risk factors early, and then provide dietary advice tailored to the mother.
The Singapore-Gravida study is co-funded by a grant from Singapore's Agency of Science, Technology and Research and New Zealand's Ministry of Business, Innovation and Employment.
Chemical fingerprints
Our individual metabolomics profile in blood and hair is almost unique, like a fingerprint of chemicals.
But profiles of women who have particular diets, and who have pregnancy complications, may show common chemicals.
If they do, a team of researchers will build a complex mathematical model to see if what they ate had any commonality.
The computer modelling work will be conducted by a bioinformatics specialist team led by principal investigator Dr Neerja Karnani and Dr Gerard Wong, a research fellow at the Singapore Institute for Clinical Studies.
