8.2 Activity: geNomad

8.2.1 Purpose

The goal of this activity is to identify plasmid and virus sequences with geNomad. geNomad is a modern classification tool that uses a dataset of over 200,000 gene markers specific to chromosomes, plasmids, or viruses to quickly find plasmids and viruses in metagenomic sequences. Additionally, geNomad uses a deep neural network approach that is gene-independent and reference alignment-independent to classify sequences as plasmids of viruses. Starting with contigs as your input, geNomad can:

  • Identify sequences of plasmids and viruses
  • Functionally annotate plasmid and viral genes
  • Taxonomically classify viral sequences

8.2.2 Learning Objectives

  1. Run the geNomad tool in Galaxy
  2. Examine geNomad output (plasmids and virus)
  3. Plot geNomad output (plasmid length and number of genes) in RStudio on Galaxy

8.2.3 Activity 1 – Run geNomad

Estimated time: 10 min

8.2.3.1 Instructions

  1. Import the dataset corresponding to assembled contigs from PacBio-sequenced Zymo gut standard D6331 - a 3.4 Gb data subset assembled with Flye tool.
  2. Go to Tools and find geNomad
    1. Agree with geNomad license
    2. Use assembled contigs (from Flye assembly) as input
    3. Under filtering presets, select Manual settings which correspond to the default geNomad settings
    4. Click Run Tool

Image test

8.2.3.2 Questions

1. Provide your Galaxy history with geNomad output:

2. Based on geNomad output file names, what information does geNomad provide?

8.2.4 Activity 2 – Explore geNomad output in Galaxy

Estimated time: 10 min

8.2.4.1 Instructions

  1. Import two of eight geNomad output files for Zymo gut standard D6331 (Activity 1 output):
  2. View geNomad output files to explore.

8.2.4.2 Questions

  1. Click on geNomad-plasmid-summary file and answer the questions below:
A. How many contigs were classified as plasmid-derived contigs?

B. Does every plasmid contig have a conjugation gene?

C. Does every plasmid contig have an amr_gene (anti-microbial resistance gene)?

  1. Click on geNomad-virus-summary file and answer the questions below:
A. How many contigs were classified as virus-derived contigs?

B. Google search full taxonomy of the first viral contig in the file and record below 1 thing you learned about this virus.

C. What is the most common genomic context (topology) in which virus was identified in this data subset?

8.2.5 Activity 3 – Examine geNomad plasmids in RStudio

Estimated time: 15 min

8.2.5.1 Instructions

  1. Launch RStudio tool in Galaxy
    • Click on “Interactive Tools” in the left hand Activity Bar and launch RStudio
    • You don’t need to include input datasets with your RStudio launch - we will import data once RStudio is launched.

Image test

  1. Import data into RStudio

    1. In your Galaxy history, identify which Galaxy history number (dataset) corresponds to the plasmid summary output file.

      • Let’s assume dataset 41 in your Galaxy history is a plasmid summary file from geNomad and you want to read it into your RStudio.

    2. In your RStudio Console, use the function gx_get() to import (copy) a dataset of interest from Galaxy history to your RStudio session.

      # Get Galaxy history dataset #41
gx_get(41)

    3. In addition, you have to use a proper R function to read the file. To read tabular files formatted as tab-separated values (tsv), use function read_tsv(). To do so, you will first need to load an R package called tidyverse. Use the following pieces of code:

      # Load tidyverse
library(tidyverse)

# Read a .tsv file
read_tsv(gx_get(41))

    4. Now that you have all pieces of code, save your tsv file as an object called, e.g. plasmids (or give it another convenient name of your choice).

      # Final import command
plasmids <- read_tsv(gx_get(41))

    5. Once code is ready, type the 2 commands (to load tidyverse, and to import plasmid_summary tabular file) into your RStudio console

Image test

  1. Create an R script of the commands from Step 2:
    • Just like the best practice for wet-lab experiments is keeping a lab notebook, the best practice for computational experiments is keeping a notebook with all your code (commands) - e.g. having a record of your R script.
    • For best practice, you should annotate each block of code

Image test

Image test

8.2.5.2 Questions

8.2.5.3 Part 1 - Explore data in RStudio

  1. Preview first rows of a file using function head().
Command: head(plasmids)
Question: Copy and paste output of command below:
Answer:
  1. View column names using function colnames().
Command: colnames(plasmids)
Question: Copy and paste output of command below:
Answer:
  1. Summarize each variable using funciton summary().
Command: summary(plasmids)
Question: Copy and paste output of command below:
Answer:
  1. Check the number of rows and columns - dimensions - of the file using function dim().
Command: dim(plasmids)
Question: Copy and paste output of command below:
Answer:

8.2.5.4 Part 2 - Plot histogram of plasmid lengths

  1. Plot histogram of plasmid lengths using function hist().
    • Note, a dollar sign ($) in the command below indicates column
Command: hist(plasmids$length)
Question: Copy and paste output of command below:
Answer:

8.2.5.5 Part 3 - Plot histogram of plasmid gene density

  1. Plot histogram of plasmid gene density (n_genes) using function hist().
    • Note, a dollar sign ($) in the command below indicates column
A. Type your command below:

B. Paste your output histogram below::

8.2.6 Grading Criteria

  • Download as Microsoft Word (.docx) and upload on Canvas

8.2.7 Footnotes

Resources

Contributions and Affiliations

  • Valeriya Gaysinskaya, Johns Hopkins University
  • Frederick Tan, Johns Hopkins University

Last Revised: March 2026

All materials CC-BY unless noted otherwise.