niupuhua1234/trimer

TRIMER is a package for building integrated metabolic–regulatory models base on Bayesian network. TRIMER can be used for knockout phenotype prediction and knock flux prediction.

---

Welcome to TRIMER Library

TRIMER is a package for building genome-scale integrated metabolic–regulatory models based on Bayesian network. The integrated model can be used for tasks such as knockout phenotypes and knockout flux prediction.

Table of Contents

• Introduction
• Dependency
• Usage
• Citation

Introduction

Transcriptional regulation plays a key role in controlling metabolism and a forefront challenge in modeling organisms today is to build integrated models of regulation and metabolism. Predicting the effect of transcriptional regulations on the metabolic network can lead to accurate predictions on how genetic mutations and perturbations are translated into flux responses at the metabolic level. TRIMER enables the quantitative integration of regulatory and metabolic networks to build genome-scale integrated metabolic–regulatory models. TRIMER is a Bayesian extension of PROM (Probabilistic Regulation of Metabolism).In TRIMER, transcriptional reguation on the metabolic network is represented by BN(Bayesian Network).

The construction of an integrated metabolic-regulatory network using TRIMER requires the following: 1) the reconstructed genome scale metabolic network 2) regulatory network structure, consisting of transcription factors (TF) and their targets 3) gene expression data. We used TRIMER to build genome-scale models for various model organisms and showed that TRIMER can identify gene knockout phenotypes with accuracies as high as 95% and predict microbial growth-rates of transcription factor knockout strains quantitatively with correlation of 0.96.

Structure of the code

Prerequisites

1. CPLEX:Detail about calling CPLEX function in matlab can be found in CPLEX official website.
2. GLPK: We suggest using the GLPK solver in COBRA package as matlab interface are provided.You can either install COBRA or copy the GLPK package to the TRIMER folder.

Usage

1. Required input data for Bayesian network learning and flux prediction.

   Gene Expression :we used the expression dataset in EcoMAC.For convenience,the raw gene expression data and binarized gene expression data can be found in raw_dataand bin data.

   Interaction List :we used the interaction list in EcoMAC which are converted form RegulonDB 8.1.

   Metabolic Model :we used iAF260 model in .mat format. iAF1260 is a metabolic model for E.coli.

   Boolean Network : we use imc1010 which is a boolearn regulatory network for E.coli.

   Interaction list , metabolic model and boolean network we used are already saved in folder source_data for convenience.

2. (Under Any R environment) Run R script shown below for Bayesian network learning before estimating the regulatory bound for flux prediction. The BN learning is seperated from other part of the package as the process is time-consuming. The input are binarized gene expression data and interaction list which serve as prior knowledge for structure leaning. The learned BN is saved in .bif format which can be read by function read.bif in bnlearn package.

   bnlearn.R
   

3. (Under Matlab Environment) The two matlab scripts shown below are demo codes of knock-out flux predicton for indole and biomass.

   flux_indole.m
   flix_biomass.m
   

   The following matlab script is a demo code for phenotypes prediction.The Tiger-Trimer model used in the demo code are already saved in source_data folder.

   prediction_phenotype.m
   

Citation

If you find our library useful, please considering citing our paper in your publications. We provide a BibTeX entry below.

@article{niu2021trimer,
  title={TRIMER: Transcription regulation integrated with metabolic regulation},
  author={Niu, Puhua and Soto, Maria J and Yoon, Byung-Jun and Dougherty, Edward R and Alexander, Francis J and Blaby, Ian and Qian, Xiaoning},
  journal={Iscience},
  volume={24},
  number={11},
  pages={103218},
  year={2021},
  publisher={Elsevier}
}

@article{niu2022protocol,
  title={Protocol for condition-dependent metabolite yield prediction using the TRIMER pipeline},
  author={Niu, Puhua and Soto, Maria J and Yoon, Byung-Jun and Dougherty, Edward R and Alexander, Francis J and Blaby, Ian and Qian, Xiaoning},
  journal={STAR protocols},
  volume={3},
  number={1},
  pages={101184},
  year={2022},
  publisher={Elsevier}
}