Taxonomy: Kraken2

Overview

After assembly quality control, we can classify the assembled genome to check the likely taxonomic identity of the sample.

In this lesson, we will use Kraken2 to classify the filtered assembly file and Krona to create an interactive taxonomic visualization.

Learning objectives

By the end of this lesson, you should be able to:

• create a folder for taxonomic classification results
• activate the classify environment
• run Kraken2 on an assembled genome
• generate a Kraken2 taxonomic report
• convert the Kraken2 report into Krona format
• create and download an interactive Krona HTML plot

Input file

In the previous assembly QC lesson, we created a filtered assembly file:

assemblies/S1_contigs_filtered.fasta

Check that the file exists:

ls assemblies

You should see:

S1_contigs_filtered.fasta

NoteImportant

For this lesson, we will use the filtered assembly file:

assemblies/S1_contigs_filtered.fasta

Create a classification directory

Create a directory named classify:

mkdir -p classify

Check that the directory was created:

ls

Activate the classify environment

Activate the classify environment:

conda activate classify

Check that kraken2 is available:

kraken2 --version

Check that the Krona tools are available:

ktImportText

If the command prints help text or usage information, Krona is available.

Run Kraken2

Run Kraken2 on the filtered assembly:

kraken2 \
  --db ~/databases/kraken2db \
  --threads 4 \
  --report classify/S1.kraken2.report \
  --output classify/S1.kraken2.output \
  assemblies/S1_contigs_filtered.fasta

Understanding the Kraken2 command

Option	Meaning
kraken2	Runs Kraken2
--db ~/databases/kraken2db	Path to the Kraken2 database
--threads 4	Use 4 CPU threads
--report classify/S1.kraken2.report	Summary taxonomic report
--output classify/S1.kraken2.output	Detailed classification output
assemblies/S1_contigs_filtered.fasta	Input assembly file

TipKraken2 command with memory issue

If you see errors in terminal mentioning that Kraken2 cannot run due to memory insufficiency, use the flag --memory-mapping to run efficiently

kraken2 \
  --db ~/databases/kraken2db \
  --memory-mapping \
  --threads 4 \
  --report classify/S1.kraken2.report \
  --output classify/S1.kraken2.output \
  assemblies/S1_contigs_filtered.fasta

Check Kraken2 outputs

List the classify directory:

ls classify

You should see:

S1.kraken2.output
S1.kraken2.report

View the first few lines of the report:

head classify/S1.kraken2.report

NoteS1.kraken2.report

100.00  80  0   R   1        root
100.00  80  0   R1  131567     cellular organisms
100.00  80  0   D   2            Bacteria
100.00  80  0   K   3379134        Pseudomonadati
100.00  80  0   P   1224             Pseudomonadota
100.00  80  0   C   1236               Gammaproteobacteria
100.00  80  0   O   135623               Vibrionales
100.00  80  0   F   641                    Vibrionaceae
100.00  80  6   G   662                      Vibrio
 92.50  74  65  S   666                        Vibrio cholerae

The Kraken2 report contains taxonomic classification information, including the percentage of sequences assigned to different taxa.

Kraken2 report columns

A Kraken2 report usually contains columns similar to:

Column	Meaning
Percentage	Percentage of sequences assigned to the taxon
Clade reads	Reads/sequences assigned to the taxon and its descendants
Taxon reads	Reads/sequences assigned directly to the taxon
Rank code	Taxonomic rank
NCBI taxonomic ID	Taxonomy ID
Scientific name	Taxon name

NoteQuestion

Which genus or species has the highest assignment?

In this example, most classified sequences belong to:

Vibrio cholerae

This suggests that sample S1 is likely Vibrio cholerae.

NoteQuestion

What is the main species-level taxonomic assignment in this report?

TipAnswer

The main species-level assignment is:

Vibrio cholerae

It has 92.50% assignment in this example.

Create Krona input file

To visualize the Kraken2 report interactively, first convert it into Krona-compatible format.

kreport2krona.py \
  -r classify/S1.kraken2.report \
  -o classify/S1.krona

Check that the Krona input file was created:

ls classify

You should see:

S1.krona

Create Krona HTML plot

Now create an interactive Krona HTML plot:

ktImportText \
  -o classify/S1.krona.html \
  classify/S1.krona

Check that the HTML file was created:

ls classify

You should see:

S1.krona.html

Windows

Open S1.krona.html manually from File Explorer using a web browser.

macOS

Use:

open S1.krona.html

Linux

Use:

xdg-open S1.krona.html

![Krona output]

Compare Kraken2 report and Krona plot

The Kraken2 report is useful for checking exact taxonomic assignments in text format.

The Krona plot is useful for interactively exploring the taxonomic hierarchy.

Output	Use
S1.kraken2.report	Text summary of taxonomic classification
S1.kraken2.output	Detailed sequence-level classification output
S1.krona	Intermediate Krona input file
S1.krona.html	Interactive Krona visualization

NoteQuestion

Compare the Kraken2 report and Krona plot.

Do they suggest the same main organism?

TipSuggested answer

Both outputs should point to the same main organism if the classification is consistent. The Kraken2 report gives the detailed text summary, while Krona provides an interactive visualization of the same taxonomic information.

Organize classification results

Your working directory should now contain:

classify

Check the folder:

ls classify

Expected files:

S1.kraken2.output
S1.kraken2.report
S1.krona
S1.krona.html

Deactivate the classify environment:

conda deactivate

Final directory structure

At the end of this lesson, your directory should look like this:

.
├── assemblies
│   ├── S1_contigs.fasta
│   └── S1_contigs_filtered.fasta
└── classify
    ├── S1.kraken2.output
    ├── S1.kraken2.report
    ├── S1.krona
    └── S1.krona.html

Practical exercise

Complete the following tasks for sample S1:

1. Create the classify directory.
2. Activate the classify environment.
3. Run Kraken2 on S1_contigs_filtered.fasta.
4. Open the Kraken2 report.
5. Convert the Kraken2 report to Krona format.
6. Create a Krona HTML plot.
7. Download the Krona HTML file to your local computer.
8. Open the Krona plot in a browser.

CautionExercise: taxonomy summary

Create a small table in your notes:

Sample	Main taxonomic assignment	Evidence from Kraken2	Evidence from Krona
S1

Fill in the table using S1.kraken2.report and S1.krona.html.

Key points

NoteImportant

• Kraken2 can classify sequences using a reference taxonomy database.
• The input for this lesson is the filtered assembly file.
• The Kraken2 report gives a text summary of taxonomic classification.
• Krona creates an interactive HTML visualization.
• Download the .html Krona file to visualize it on your local computer.

Next step

In the next lesson, we will continue downstream analysis using the filtered assembly file.