Have you ever wondered why do we combine ingredients in our recipes the way we do? Or for that matter, could we find scientific ways for altering diet to improve health? Our data-driven investigations aimed at probing patterns in traditional Indian recipes, in response to the first question, have led us to the discovery of a novel food pairing phenomenon in Indian cuisine. Our studies have revealed ‘culinary fingerprints’ of regional cuisines and role of spice as the molecular fulcrum of Indian recipes. Such data-driven explorations of food are opening new avenues for development of divergent applications in the domains of nutrition and health. One such direction is towards application of machine learning for ‘personalized nutrition’ that can potentially answer the second question, allowing us to leverage food as medicine.
5. Why do we eat what we eat?
Why do we combine ingredients in
our recipes the way we do?
6. Cuisines: Traditional Recipes
A data and hypothesis oriented approach to food
The Molecular Essence
The quintessential molecular character
Exploiting patterns for health
Data-driven personalized food recommendation
9. 2543 Traditional Indian Recipes (TarlaDalal)
Regional cuisines: Bengali, Gujarati, Jain, Maharashtrian,
Mughlai, Punjabi, Rajasthani, South Indian.
10. Why do we combine ingredients in our recipes
the way we do?
11. Cuisines: Traditional Recipes
A data and hypothesis oriented approach to food
The Molecular Essence
The quintessential molecular character
Exploiting patterns for health
Data-driven personalized food recommendation
12. Food Pairing Hypothesis
Ingredients sharing flavor compounds are more likely to taste well
together than ingredients that do not.
𝑁𝑠 = 𝐹𝑖 ∩ 𝐹𝑗
𝐹𝑖
𝐹𝑗
Ahn et. al, “Flavor network and the principles of food pairing”, 1, 196, Scientific Reports (2011).
A Jain, NK Rakhi, G Bagler*, “Spices form the basis of food pairing in Indian cuisine”, arXiv:1502.03815 (2015).
17. Indian cuisine is
characterized with
contrasting food pairing.
more the extent of flavor sharing between any two
ingredients, lesser their co-occurrence
18. Contrasting Food Pairing
—at the level of ingredient pairs—
A Jain, NK Rakhi, G Bagler*, “Spices form the basis of food pairing in Indian cuisine”, arXiv:1502.03815 (2015).
19. A Jain, NK Rakhi, G Bagler*, “Spices form the basis of food pairing in Indian cuisine”, arXiv:1502.03815 (2015).
Contrasting Food Pairing
—at the level of cuisine—
20. A Jain, NK Rakhi, G Bagler*, “Spices form the basis of food pairing in Indian cuisine”, arXiv:1502.03815 (2015).
Contrasting Food Pairing
—at the level of sub-cuisines—
21. Indian cuisine is
characterized with
contrasting food pairing.
more the extent of flavor sharing between any two
ingredients, lesser their co-occurrence
22. A Jain, NK Rakhi & G Bagler,* “Analysis of Food Pairing in Regional Cuisines of India”, PLoS ONE, 10(10): e0139539(2015).
Culinary Fingerprinting of Regional Cuisines of India
23. A Jain, NK Rakhi, G Bagler*, “Spices form the basis of food pairing in Indian cuisine”, arXiv:1502.03815 (2015).
Spices are key to the food pairing in Indian cuisine
24. Positive (Uniform)Negative (Contrasting)
A Jain, NK Rakhi, G Bagler*, “Spices form the basis of food pairing in Indian cuisine”, arXiv:1502.03815 (2015).
Role of ingredients in biasing the food pairing
27. Discovery of the molecular essence of Indian cuisine & applications
Highlighted as an
Emerging Technology in
A Jain, NK Rakhi and G Bagler*, arXiv (2015); A Jain, NK Rakhi and G Bagler*, PLoS ONE (2015).
Best of 2015
MIT Technology Review
28.
29.
30. Cuisines: Traditional Recipes
A data and hypothesis oriented approach to food
The Molecular Essence
The quintessential molecular character
Exploiting patterns for health
Data-driven personalized food recommendation
34. Can we find scientific ways for altering diet
to improve health?
Personalized food recommendation
35. Levels of glucose in the blood are measured in terms of
“Postprandial Glycemic Response (PPGR)”.
Image Credits: HealthClinic
36.
37. Machine learning applied to multidimensional data
for personalized dietary recommendation
ED Sonnenburg and JL Sonnenburg, Nature, 528, 484 (Dec 2015).
Zeevi et al., “Personalized Nutrition by Prediction of Glycemic Responses”, Cell, 163, 1079-1094 (Nov 2015);
38. The discovery of a new dish confers
more happiness on humanity, than the
discovery of a new star.
Jean Anthelme Brillat-Savarin
“
”
as seen in IUCAA canteen circa 1998