ansalaza / ptolemy2

A stand-alone method for constructing gene-graphs

Ptolemy2

Ptolemy2 is a stand-alone method for constructing gene-graphs given long-read alignments to one or more reference genomes or pairwise whole-genome alignments.

The gene-graphs can then be used to analyze genome architecture--organization and collection of genes in a genome--across a (metagenomic) sequencing dataset or a collection of genome assemblies.

This work is an extension of our previous published method, Ptolemy, and has thus far only been tested on genome assemblies and long-read metagenomic sequencing data of bacteriphages. But feel free to try it on on other organisms!

How it works (in a nutshell...)

Requirements

Java 1.8 or higher.

Optionally, minimap2 accessible from you PATH environment.

How to run

Given long-read alignments in PAF-format:

java -jar ptolemy2.jar build-db gene-graph -g all_gffs.txt \
--feature-types "CDS" \
--name-tag "product=" \
-r read_alignments.paf \
--prefix <some_string>

For whole-genome alignments, switch '-r' parameter with '-w'.

Output is a GFA-formatted file.

You can visualize the graph with Bandage or use Ptolemy2's internal GFA-to-CSV converter and visualize with other graph visualizers like Gephi.