FDMine_Framework

Objective & Result

Food-drug interactions (FDIs) arise when nutritional dietary consumption regulates biochemical mechanisms involved in drug metabolism. This study proposes FDMine, a novel systematic framework that models the FDI problem as a homogenous graph. Our dataset consists of 788 unique approved small molecule drugs with metabolism-related drug-drug interactions and 320 unique food items, composed of 563 unique compounds. The potential number of interactions is 87,192 and 92,143 for disjoint and joint versions of the graph. We defined several similarity subnetworks comprising food-drug similarity, drug-drug similarity, and food-food similarity networks. A unique part of the graph involves encoding the food composition as a set of nodes and calculating a content contribution score. To predict new FDIs, we considered several link prediction algorithms and various performance metrics, including the precision@top (top 1%, 2%, and 5%) of the newly predicted links. The shortest path-based method has achieved a precision of 84%, 60% and 40% for the top 1%, 2% and 5% of FDIs identified, respectively. We validated the top FDIs predicted using FDMine to demonstrate its applicability, and we relate therapeutic anti-inflammatory effects of food items informed by FDIs. FDMine is publicly available to support clinicians and researchers.},

Figure: The framework of FDMine

Figure: Comparison of the precision@top over eight methods and two different graph networks

Framework Execute

Use run.sh to run the FDMine Framework. Before executing run.sh you have to change the path address in the run.sh file. You do not have to do anything else!

Dataset Description

The FDMine need 3 datasets (in .CSV formate) to execute. They are Drug_SSP.csv, Food_SSP.csv, and food_contribution.csv. The structure of the dataset are given below

Drug_SSP.csv

drugbank.id	smiles
DB01059	OC(=O)C1=CN=CC=C1
DB04908	FC(F)(F)C1=CC(=CC=C1)N1CCN(CCN2C(=O)NC3=CC=CC=C23)CC1

Food_SSP.csv

public_food_id_compound_id_name	compound_SMILES
FOOD00004 _ FDB001014 _ Nicotinic acid	CCN1C=C(C(O)=O)C(=O)C2=CC(F)=C(C=C12)N1CCNCC1
FOOD00004 _ FDB003513 _ Calcium	[Ca++]

food_contribution.csv

public_food_id_compound_id_name	contribution
FOOD00004 _ FDB001014 _ Nicotinic acid	0.000417143789774213
FOOD00004 _ FDB003513 _ Calcium	0.0246710755666463

Parameter passing from the bash script

Change the argument given in the run.sh file, if you want to set your own Tanimotot threshold and Food contribution threshold. The Tanimoto threshold is related to calculate_ssp.py and the food contribution threshold is related to food_contribution_threshold.R files respectively.

Citation

Please cite the following publication if any of the results in this paper or code are beneficial in your research:

@Article{Rahman2022,
author={Rahman, Md. Mostafizur and Vadrev, Srinivas Mukund and Magana-Mora, Arturo and Levman, Jacob and Soufan, Othman},
title={A novel graph mining approach to predict and evaluate food-drug interactions},
journal={Scientific Reports},
year={2022},
month={Jan},
day={20},
volume={12},
number={1},
pages={1061},
issn={2045-2322},
doi={10.1038/s41598-022-05132-y},
url={https://doi.org/10.1038/s41598-022-05132-y}
}

Required R Packages

• library(tidyverse) [1.3.0]
• library(igraph) [1.2.6]
• library(LinkPrediction) [1.0]
• library(bayestestR) [0.8.2]
• library(zoo) [1.8-9]
• library(dplyr) [1.0.5]

Required Python Packages

• Python==3.8.5
• rdkit==2021.03.2
• panda