Genome Assembly, Annotation, and Metabolic Pathway Prediction

Background

Microbial communities of bacteria form complex interactions in nature. However, our understanding of these communities that reside in humans and other animals is limited to those that can be cultured in the lab. In comparison to other microbiomes such as ones in soil or in the human gut, the human urinary tract (UT) harbors a lower biomass. Thus, this decreased diversity of the UT serves as an ideal model to delve into its interconnected nature, which has yet to be fully uncovered. This microbiome, which was previously thought to be sterile, hosts a diverse assortment of bacteria and other microbes that make up what is known as the urobiome. Enhanced culture methods have enabled the isolation of numerous fastidious species from the UT, definitively proving that the urinary tract of asymptomatic individuals is not sterile. To examine urobiome communities and their potential interactions metabolically, I computationally predicted metabolic pathways from bacterial genomes of a given patient's urine sample to determine if specific bacteria form dependent relationships in the urobiome for metabolic benefits.

Overview

Bacterial genomes from patients that have multiple bacterial isolates sequenced were assembled using SPAdes v 3.15.2 via the Pathosystems Resource Integration Center (PATRIC) v. 3.6.12. Genome assemblies were also annotated using PATRIC. The features.txt file produced from each annotation was parsed via Python v.3.8.5 to extract the enzyme codes (EC) needed to generate metabolic maps using KEGG Mapper-Color; a web tool for metabolic pathway prediction.

Software

• Python3
• KEGG Mapper-Color

Genomes annotated

Sample ID	Patient ID	Species
512	RUTISD6	Staphylococcus epidermidis
610	RUTISD6	Klebsiella pneumoniae
946	RUTISD6	Klebsiella pneumoniae
612	RUTISD6	Lactobacillus gasseri
615	RUTISD6	Lactobacillus delbrueckii
617	RUTISD6	Enterococcus faecalis
441	RUTISD9	Klebsiella pneumoniae
442	RUTISD9	Enterococcus faecalis
443	RUTISD9	Staphylococcus epidermidis
444	RUTISD9	Streptococcus agalactiae
445	RUTISD9	Klebsiella pneumoniae
446	RUTISD9	Lactobacillus jensenii
479	RUTISD018	Escherichia coli
102	RUTISD018	Enterococcus faecalis
482	RUTISD018	Lactobacillus jensenii
483	RUTISD018	Lactobacillus gasseri
530	RUTISD25	Escherichia coli
531	RUTISD25	Streptococcus anginosus
532	RUTISD25	Actinotignum sanguinis
533	RUTISD25	Aerococcus urinae
535	RUTISD25	Aerococcus sanguinicola
536	RUTISD25	Escherichia coli

Run Script

Clone Repository:

https://github.com/zgb963/kegg_pathway_prediction.git

Move Into Project Directory:

cd kegg_pathway_prediction/

Run script:

python3 metabolicmaps.py

Output

Each annotation .features.txt file with _results attachment. Results file contains enzyme codes and color. Copy and paste contents of file into KEGG Mapper-Color to see resulting metabolic maps. Below is an example.