Mutational signature dynamics shaping the evolution of oesophageal adenocarcinoma

Description

This code has been developed to investigate the prevalence and dynamics of mutational signatures across Barrett Oesophagus, primary tumours and metastases in oesophageal adenocarcinoma.

System Requirements

Operating system(s): Unix (Linux, Mac OS) Programming Language: R 4.1.2

Package versions utilised:

desconstructSigs v1.9.0

MutationalPatterns v3.10.0

MutationTimer v1.00.0

dNdScv v0.1.0

xgboost v1.7.3.1

randomForest v4.7-1.1

glmnet v4.1-7

Installation guide

Make sure you are running R version 4.1.2 or above, and install the required package dependencies as indicated in the System Requirements section above.

Typical install time on a "normal" desktop computer: negligible, mostly reliant on dependent packages.

Data

The data employed for this analysis can be found in the /data folder, the Source Data file of the corresponding paper or are available upon request.

Code

The scripts employed in this analysis can be found in the /code folder.

MutationalSignatureAnalysis:

Scripts for the inference of mutational signatures and their clonality.

PositiveSelectionAnalysis:

Positive selection analysis using the dndscv method for defined subgroups of the cohort.

AllOtherAnalyses:

Script employed to generate the results and figures presented in the paper, numbered according to the figures in the corresponding paper. They include the following analyses:

- Linking signatures with genomic markers and pathway activities:

2_compareSignaturesStages.R

4_oncoprintDrivers_sigcutoff05.R

6_CINandS17.R

6_analyseClonality.tracksig.R

6_signaturePresenceCorrWithProgrammes.R

s1_TMB.R

s21_upsetPlots.R

s22-23_PIK3CA_KRAS_APC.R

s24_clonalityPurity.R

s27_signature17.R

s3_compareSignaturesStages_indels.R

- Gradient boost and multinomial regression classifiers of disease stages based on mutational signatures:

3_multipleLinearReg-oct2021.R

3_xgboost_withcv_april2022.R

s25_xgboost_withcv_jan2023_plusindels.R

s26_final_xgboostsimpleWithClonality_primvsmets.R

- DDR process analyses:

5_ddr.snvs.2023.R

5_genericDDRpatterns.R

7_SBS30_BER.R

s4-5_MMRandImmunity.R

- Investigating the relation between mutational signatures and clinical outcomes:

7_linkTreatmentResponse.R

s14_produceClinicalPlot.R

s15-19_clinicalAssociations_smoking.R

s6_barretts_heterogeneity.R

- Strelka vs Mutect comparison:

s29_mutectVSStrelka.R

Demo / Instructions for use

To run a particular script, navigate to the corresponding folder and run the script using the command Rscript name_of_script.R. Outputs include results of statistical tests, plots and newly generated result files.

Expected run times: Most scripts should run instantaneously, except for the MutationTimer analysis that will depend on the size of your input dataset. The mutation signature-based classifiers can take 2-5 minutes to run on average.

Publication

This code has been generated for the following publication:

Mutational signature dynamics shaping the evolution of oesophageal adenocarcinoma. (2023) Sujath Abbas, Oriol Pich, Ginny Devonshire, Shawn A. Zamani, Annalise Katz-Summercorn, Sarah Killcoyne, Calvin Cheah, Barbara Nutzinger, Nicola Grehan, Nuria Lopez-Bigas, OCCAMS Consortium, Rebecca C. Fitzgerald*, Maria Secrier*

Copyright

This code is free and is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY. See the GNU General Public License for more details.