Multiplexed assays of variant effect (MAVEs), including deep mutational scanning (DMS) experiments, measure the functional consequences of many genetic variants in parallel. Instead of testing variants one at a time, a library of variants is assayed in a pooled experiment. Sequencing before and after selection then estimates how each variant changed in frequency during the assay.

In this tutorial, we will use a CHEK2 variant effect dataset from MaveDB (urn:mavedb:00001203-a-2) to calculate a functional score with CountESS. The original study tested coding single nucleotide variants (SNVs) in the human CHEK2 open reading frame. Loss-of-function mutations in CHEK2, which encodes the checkpoint kinase CHK2, are associated with increased risk of breast and other cancers. Many CHEK2 variants observed in clinical testing are variants of uncertain significance, so functional assays can provide evidence for interpreting their possible impact. In this study, variants were tested for their ability to complement a Saccharomyces cerevisiae strain lacking the yeast ortholog RAD53 McCarthy-Leo et al. 2024. The resulting RAD53 Complementation Score (RCS) describes whether a CHEK2 variant was depleted or retained after selection.

The training data for this tutorial starts from a preprocessed variant frequency summary, not raw sequencing reads. In the published analysis, reads were quality filtered and mapped to the CHEK2 amplicon, variants were detected, and each variant frequency was calculated as the number of reads containing that variant divided by the sequencing coverage at that nucleotide position. Those frequencies were averaged across the two pre-selection library aliquots and across the four post-selection screens. The frequency values were multiplied by 1000 to make small variant frequencies easier to read and compare. More details are available in the original publication McCarthy-Leo et al. 2024. The two frequency columns used here are:

• Ave_LibAandB_x1000: average variant frequency before selection, from the two input library aliquots.
• Ave_S1toS4_x1000: average variant frequency after selection, from four complementation screens.

CountESS will calculate the log-ratio score:

RCS_this_SNV = log2(Ave_S1toS4_x1000 / Ave_LibAandB_x1000)

Negative scores indicate variants depleted after selection, consistent with impaired CHEK2 function in this assay. Scores near zero indicate variants with similar frequencies before and after selection, suggesting little detectable effect in this assay. Positive scores indicate variants that increased in frequency after selection and were retained better than the average variant population.

Agenda

In this tutorial, we will cover:

1. Data preparation
2. Inspect the input files
3. Run the CountESS workflow
4. Interpret the output
5. Visualize the score distribution
6. Compare with MaveDB
7. Conclusion

Data preparation

The CountESS workflow and input data are available from Zenodo.

Hands On: Data upload

1. Create a new history for this tutorial.

   Tip: Creating a new history

   To create a new history simply click the new-history icon at the top of the history panel:

   

2. Import the following files from Zenodo:

   https://zenodo.org/records/20250971/files/chek2_frequency_summary.csv
   https://zenodo.org/records/20250971/files/chek2_rcs_countess.ini
   

   Tip: Importing via links

   • Copy the link location

   • Click galaxy-upload Upload at the top of the activity panel

   • Select galaxy-wf-edit Paste/Fetch Data

   • Paste the link(s) into the text field

   • Press Start

   • Close the window

   

3. Set the datatype of chek2_frequency_summary.csv to csv if Galaxy does not detect it automatically.

   Tip: Changing the datatype

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
   • In the central panel, click galaxy-chart-select-data Datatypes tab on the top
   • In the galaxy-chart-select-data Assign Datatype, select csv from “New Type” dropdown
     • Tip: you can start typing the datatype into the field to filter the dropdown menu
   • Click the Save button

4. Set the datatype of chek2_rcs_countess.ini to txt if Galaxy does not detect it automatically.

   Tip: Changing the datatype

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
   • In the central panel, click galaxy-chart-select-data Datatypes tab on the top
   • In the galaxy-chart-select-data Assign Datatype, select txt from “New Type” dropdown
     • Tip: you can start typing the datatype into the field to filter the dropdown menu
   • Click the Save button

Inspect the input files

The input CSV contains one row per CHEK2 variant and includes variant identifiers, HGVS-style variant descriptions, and frequency summaries from before and after selection.

Question

1. Which two columns will be used to calculate the RCS score?

Solution

The workflow uses Ave_LibAandB_x1000 and Ave_S1toS4_x1000.

The second input file, chek2_rcs_countess.ini, is a CountESS configuration file. It is a saved CountESS workflow: each section in the .ini file describes one workflow node, and the _parent lines describe how nodes are connected.

The saved workflow has three CountESS nodes:

1. Load CHEK2 frequencies: load the CSV input table.
2. Calculate RCS: calculate the log-ratio score.
3. Save CHEK2 RCS scores: write the output CSV.

This workflow is intentionally small so that the score calculation is easy to inspect. CountESS workflows created in the GUI can be more complex, and the Galaxy tool can run those saved workflows too.

Details: How was the CountESS configuration file created?

The provided .ini file was created in the CountESS GUI. CountESS GUI is external software: to create or edit this configuration yourself, CountESS needs to be installed and the GUI needs to be run outside Galaxy first. You do not need to recreate the .ini file to complete this tutorial, but the workflow is intentionally simple:

1. Add a CSV Load node, rename it to Load CHEK2 frequencies, and load chek2_frequency_summary.csv.

   Open image in new tab
   
   

2. Add an Expression node connected to the CSV Load node and rename it to Calculate RCS.

3. In the Expression node, enter:

   RCS_this_SNV = log2(Ave_S1toS4_x1000 / Ave_LibAandB_x1000)
   

   Open image in new tab
   
   

4. Add a CSV Save node connected to the Expression node, rename it to Save CHEK2 RCS scores, and set the output filename to chek2_rcs_scores.csv.

   Open image in new tab
   
   

5. Save the CountESS configuration as chek2_rcs_countess.ini.

Run the CountESS workflow

CountESS is a workflow-like table processing tool. Galaxy will run this saved CountESS workflow, connect Galaxy datasets to the input nodes, and collect files written by CountESS save nodes into an output collection.

Hands On: Calculate CHEK2 RCS scores with CountESS

1. Run CountESS ( Galaxy version 0.1.26+galaxy0) with the following parameters:

   • param-file “CountESS configuration file”: chek2_rcs_countess.ini

   • In “Input file mapping”:
     • “CountESS input node name”: Load CHEK2 frequencies
     • param-file “Input dataset”: chek2_frequency_summary.csv
   • “Log level”: INFO

CountESS produces a collection of output files from the save-node filenames stored in the .ini file. Open the output collection and inspect chek2_rcs_scores.csv.

Interpret the output

The output table contains the calculated RCS_this_SNV values. These scores are log2 ratios of the average post-selection variant frequency over the average pre-selection variant frequency.

For example, a variant with lower frequency after selection than before selection will have a negative RCS value. A variant with similar frequencies before and after selection will have an RCS close to zero. A variant with higher frequency after selection than before selection will have a positive RCS value. In the CHEK2 RAD53 complementation assay, strongly negative RCS values indicate variants that were depleted during selection and are therefore more likely to impair CHEK2 function in this assay.

Question

1. What do negative, near-zero, and positive RCS values mean in this experiment?

Solution

A negative RCS value means the variant was less frequent after selection than before selection, consistent with reduced ability of that CHEK2 variant to complement the yeast rad53 mutant. A near-zero RCS means the variant frequency changed little during selection. A positive RCS means the variant was more frequent after selection than before selection, suggesting it was retained better in this assay.

Visualize the score distribution

A simple next step is to plot the distribution of calculated RCS values. This gives a quick overview of the range of functional effects in the assay and helps identify whether most variants are close to neutral, depleted, or enriched after selection.

Hands On: Plot the distribution of calculated RCS values

1. Run Cut columns from a table with the following parameters:

   • “Cut columns”: c6
   • “Delimited by”: Comma
   • param-collection “From”: chek2_rcs_scores.csv from the CountESS output collection

2. Run Histogram with ggplot2 ( Galaxy version 3.5.1+galaxy1) with the following parameters:

   • param-collection “Input should have column headers - these will be the columns that are plotted”: output of Cut columns from a table tool
   • “Label for x axis”: RAD53 Complementation Score
   • “Label for y axis”: Number of variants
   • “Bin width for plotting”: 0.5
   • In “Advanced Options”:
     • “Plot counts or density”: Plot counts on the y-axis
     • “Legend options”: Hide legend

Question

1. What does this histogram show that is not obvious from inspecting individual rows?

Solution

The histogram summarizes the whole score distribution. Most variants have scores near zero, with a left tail of depleted variants and fewer variants with positive scores. This is useful for getting a first overview of the assay before selecting variants or score ranges for closer inspection.

Open image in new tab

Compare with MaveDB

The MaveDB score set used for this tutorial is:

https://mavedb.org/score-sets/urn:mavedb:00001203-a-2

The calculated RCS_this_SNV values from CountESS should match the RCS_this_SNV column available from MaveDB for this score set. To check this directly, we can import the MaveDB score table, join it to the CountESS output by accession, and plot the two RCS columns against each other. If the CountESS workflow reproduced the deposited log-ratio values, the points should fall on a 1:1 diagonal.

Hands On: Compare CountESS scores with the MaveDB score table

1. Import the MaveDB score table into Galaxy using this link:

   https://api.mavedb.org/api/v1/score-sets/urn:mavedb:00001203-a-2/scores
   

   Tip: Importing via links

   • Copy the link location

   • Click galaxy-upload Upload at the top of the activity panel

   • Select galaxy-wf-edit Paste/Fetch Data

   • Paste the link(s) into the text field

   • Press Start

   • Close the window

   

2. Set the datatype of the imported MaveDB score table to csv if Galaxy does not detect it automatically.

   Tip: Changing the datatype

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
   • In the central panel, click galaxy-chart-select-data Datatypes tab on the top
   • In the galaxy-chart-select-data Assign Datatype, select csv from “New Type” dropdown
     • Tip: you can start typing the datatype into the field to filter the dropdown menu
   • Click the Save button

3. Convert the imported MaveDB score table from CSV to tabular format.

   Tip: Converting the file format

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes.
   • In the central panel, click galaxy-chart-select-data Datatypes tab on the top.
   • In the galaxy-gear Convert to Datatype section, select tabular (using csv-to-tabular) from “Target datatype” dropdown.
   • Click the Create Dataset button to start the conversion.

4. Rename the converted dataset to MaveDB CHEK2 scores tabular.

   Tip: Renaming a dataset

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
   • In the central panel, change the Name field
   • Click the Save button

5. Convert chek2_rcs_scores.csv from the CountESS output collection from CSV to tabular format.

   Tip: Converting the file format

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes.
   • In the central panel, click galaxy-chart-select-data Datatypes tab on the top.
   • In the galaxy-gear Convert to Datatype section, select tabular (using csv-to-tabular) from “Target datatype” dropdown.
   • Click the Create Dataset button to start the conversion.

6. Rename the converted dataset to CountESS CHEK2 RCS scores tabular.

   Tip: Renaming a dataset

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
   • In the central panel, change the Name field
   • Click the Save button

7. Run Join two Datasets side by side on a specified field with the following parameters:

   • param-file “Join”: MaveDB CHEK2 scores tabular
   • “using column”: Column: 1
   • param-file “with”: CountESS CHEK2 RCS scores tabular
   • “and column”: Column: 1
   • “Keep lines of first input that do not join with second input”: No
   • “Keep lines of first input that are incomplete”: No
   • “Keep the header lines”: Yes

8. Rename the joined dataset to MaveDB and CountESS RCS comparison.

   Tip: Renaming a dataset

   • Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
   • In the central panel, change the Name field
   • Click the Save button

9. Run Scatterplot with ggplot2 ( Galaxy version 3.5.1+galaxy2) with the following parameters:

   • param-file “Input in tabular format”: MaveDB and CountESS RCS comparison
   • “Column to plot on x-axis”: c8: RCS_this_SNV
   • “Column to plot on y-axis”: c30: RCS_this_SNV
   • “Plot title”: MaveDB and CountESS CHEK2 RCS values
   • “Label for x axis”: MaveDB RCS_this_SNV
   • “Label for y axis”: CountESS RCS_this_SNV
   • In “Advanced Options”:
     • “Data point options”: User defined point options
       • “relative size of points”: 1.0

In the joined table, column 8 is the RCS_this_SNV value from MaveDB. Column 30 is the RCS_this_SNV value from the CountESS output, because the six CountESS output columns are added after the 24 MaveDB score-table columns.

Question

1. What does the scatterplot show about the relationship between the MaveDB and CountESS RCS values?

Solution

The points should lie on a 1:1 diagonal, showing that the RCS_this_SNV values calculated by the CountESS workflow match the RCS_this_SNV values deposited in MaveDB for the same accessions. The plot shows this expected pattern: the CountESS score on the y-axis is essentially the same as the MaveDB score on the x-axis for each variant. This confirms that the Galaxy CountESS workflow reproduced the deposited log-ratio values.

Open image in new tab

MaveDB also contains a final score column for this record. That final score is not the direct log-ratio calculated in this tutorial. It is derived downstream from the RCS values using a logistic regression model. In this tutorial, we focus on reproducing the transparent log-ratio component of the analysis.

Conclusion

In this tutorial, you used Galaxy to run a saved CountESS workflow and calculate CHEK2 RAD53 Complementation Scores from preprocessed DMS/MAVE frequency data. You then visualized the distribution of the calculated log-ratio scores.

The CountESS output can be used as input for several downstream analyses. In the original CHEK2 analysis, the RCS values were used as input to a logistic regression model to estimate the probability that variants are functionally damaging. Researchers can use these scores to compare the effects of synonymous, missense, and nonsense variants, and to identify variants or score ranges that may need closer inspection. The scores can also be combined with clinical annotations, mapped onto protein domains, or used to prioritize variants of uncertain significance for follow-up. Functional scores can be calibrated for clinical evidence using approaches such as ExCALIBR, which can help translate assay results into evidence for classifying variants as more likely benign or pathogenic.

You've Finished the Tutorial

Please also consider filling out the Feedback Form as well!

Key points

• MAVE experiments can measure functional effects for thousands of variants in parallel.

• Variant frequencies before and after selection can be transformed into functional scores using log-ratio calculations.

• CountESS workflows saved from the CountESS GUI can be reused in Galaxy for reproducible analysis.

Frequently Asked Questions

• Topic FAQs
• Galaxy FAQs
• GTN Matrix Chat
• Galaxy Help Forum

Useful literature

Further information, including links to documentation and original publications, regarding the tools, analysis techniques and the interpretation of results described in this tutorial can be found here.

References

1. McCarthy-Leo, C. E., G. S. Brush, R. Pique-Regi, F. Luca, M. A. Tainsky et al., 2024 Comprehensive analysis of the functional impact of single nucleotide variants of human CHEK2. PLOS Genetics 20: e1011375. 10.1371/journal.pgen.1011375

Feedback

Did you use this material as an instructor? Feel free to give us feedback on how it went.
Did you use this material as a learner or student? Click the form below to leave feedback.

Citing this Tutorial

1. Polina Polunina, Tristan Reynolds, Nick Moore, Alan Rubin, Calculating CHEK2 variant effect scores from MAVE data with CountESS (Galaxy Training Materials). https://training.galaxyproject.org/training-material/topics/variant-analysis/tutorials/countess-mave-chek2/tutorial.html Online; accessed TODAY
2. Hiltemann, Saskia, Rasche, Helena et al., 2023 Galaxy Training: A Powerful Framework for Teaching! PLOS Computational Biology 10.1371/journal.pcbi.1010752
3. Batut et al., 2018 Community-Driven Data Analysis Training for Biology Cell Systems 10.1016/j.cels.2018.05.012

BibTeX

@misc{variant-analysis-countess-mave-chek2,
author = "Polina Polunina and Tristan Reynolds and Nick Moore and Alan Rubin",
	title = "Calculating CHEK2 variant effect scores from MAVE data with CountESS (Galaxy Training Materials)",
	year = "",
	month = "",
	day = "",
	url = "\url{https://training.galaxyproject.org/training-material/topics/variant-analysis/tutorials/countess-mave-chek2/tutorial.html}",
	note = "[Online; accessed TODAY]"
}
@article{Hiltemann_2023,
	doi = {10.1371/journal.pcbi.1010752},
	url = {https://doi.org/10.1371%2Fjournal.pcbi.1010752},
	year = 2023,
	month = {jan},
	publisher = {Public Library of Science ({PLoS})},
	volume = {19},
	number = {1},
	pages = {e1010752},
	author = {Saskia Hiltemann and Helena Rasche and Simon Gladman and Hans-Rudolf Hotz and Delphine Larivi{\`{e}}re and Daniel Blankenberg and Pratik D. Jagtap and Thomas Wollmann and Anthony Bretaudeau and Nadia Gou{\'{e}} and Timothy J. Griffin and Coline Royaux and Yvan Le Bras and Subina Mehta and Anna Syme and Frederik Coppens and Bert Droesbeke and Nicola Soranzo and Wendi Bacon and Fotis Psomopoulos and Crist{\'{o}}bal Gallardo-Alba and John Davis and Melanie Christine Föll and Matthias Fahrner and Maria A. Doyle and Beatriz Serrano-Solano and Anne Claire Fouilloux and Peter van Heusden and Wolfgang Maier and Dave Clements and Florian Heyl and Björn Grüning and B{\'{e}}r{\'{e}}nice Batut and},
	editor = {Francis Ouellette},
	title = {Galaxy Training: A powerful framework for teaching!},
	journal = {PLoS Comput Biol}
}

                   

Funding

These individuals or organisations provided funding support for the development of this resource

ELIXIR Europe

UFR

The University of Melbourne

Melbourne Bioinformatics

Australian BioCommons

Galaxy Administrators: Install the missing tools

You can use Ephemeris's shed-tools install command to install the tools used in this tutorial.

shed-tools install [-g GALAXY] [-a API_KEY] -t <(curl https://training.galaxyproject.org/training-material/api/topics/variant-analysis/tutorials/countess-mave-chek2/tutorial.json | jq .admin_install_yaml -r)

Alternatively you can copy and paste the following YAML

---
install_tool_dependencies: true
install_repository_dependencies: true
install_resolver_dependencies: true
tools:
- name: countess
  owner: iuc
  revisions: b1d197da67aa
  tool_panel_section_label: Variant Calling
  tool_shed_url: https://toolshed.g2.bx.psu.edu/
- name: ggplot2_histogram
  owner: iuc
  revisions: 6b9816a913de
  tool_panel_section_label: Graph/Display Data
  tool_shed_url: https://toolshed.g2.bx.psu.edu/
- name: ggplot2_point
  owner: iuc
  revisions: f87634b12749
  tool_panel_section_label: Graph/Display Data
  tool_shed_url: https://toolshed.g2.bx.psu.edu/

Feedback

No feedback has been recieved yet for this training. Be the first one by filling in the feedback form.