problem statement

Metabolic engineering is the modification of an organism’s cellular activities through recombinant DNA technology to optimize the production of a target biochemical. The central limitation to metabolic engineering is the difficulty of predicting resulting reactions from genetic manipulation due to the volume, complexity, and variability of every organism's metabolic pathways.

Our project aims to develop an algorithm that will provide a ranked series of optimization gene-deletion strategies, for the model organism saccharomyces cerevisiae, from user defined target metabolic pathways.

GOALS&

OBJECTIVES

Our Goal is to develop a versatile program that outputs transcription factor knockouts that are advantageous for upregulation or downregulation of a gene. In order to meet this goal our team has the following objectives:

To extract gene and TF interaction data from the database: YEASTRACT (Yeast Search for Transcriptional Regulators And Consensus Tracking) is a curated repository of more than 206000 regulatory associations between transcription factors (TF) and target genes in Saccharomyces cerevisiae

import a matrix developed from data collected from YESTRACT into matlab

create an algorithm based on this developed matrix that gives the user the ability to input whether they want to up or downregulate a gene, and the output will be a ranked list of transcription factor knockouts to achieve their goal.

CONSTRAINTS&

CRITERIA

The model must predict and rank multiple optimization strategies that increase production of any target biochemical based on:

Minimal deletions
Viability of organism after manipulation. Will it survive after genetic manipulations?
Cellular growth rate and biochemical productivity
The regulatory network information must be complete, accurate, and encompassed in the algorithm

Linear assumptions must be made. The regulatory information is analyzed assuming the interactions between a TF and the genes hold with involvement of multiple TFs and after deletions. The TFs will not influence each other.
Limited computational power
Recombinant DNA technologies still need improvement