Large-scale genomic surveillance reveals immunosuppression drives mutation dynamics in persistent SARS-CoV-2 infections

Code for "Large-scale genomic surveillance reveals immunosuppression drives mutation dynamics in persistent SARS-CoV-2 infections"

Link to paper: To be added

Overview

The presented code aims to reproduce the analyses from "Large-scale genomic surveillance reveals immunosuppression drives mutation dynamics in persistent SARS-CoV-2 infections". In brief, the goal of the code is to identify persistent infections, identify risk factors for persistent infections, analyze associated mutational patterns, and link health conditions to accelerated selective genetic changes in the virus.

Folders

• case_control_code: Code used for preparing data and running the case control study.
• defining_snp_thresholds: Contains code to identify the false-positive rates at different rare SNP thresholds.
• dnds: Code to conduct dnds/kaks analyses.
• figures: Contains code that leads to the four main figures in the article.
• identifying_persistent_infections: Code used to identify persistent infections from the full set of sequences.
• identifying_substitutions_and_mutations: Code used to identify the placement of mutations and non-synonymous changes, as well as analyzing the association between different risk factors and diagnoses with the rate of (non)synonymous change.
• identical_sequences: Code to analyze the distance between the final persistent infection sequences and all other sequences in the database within 11 days.
• rebounding: Code to analyze the occurrence of rebounding among persistent infections.
• recurrent_mutations: Code to identify independently-occurring and recurrent mutations among persistent infections.
• tip_length_analysis: Code to analyze tip lengths for persistent versus control infections.

All code is numbered in the intended order of use.

General Data

Synthetic data is available under the data folder in figures. Due to confidentiality restrictions, dummy/synthetic data has been generated; they do not necessarily correspond to any real data, but were generated to allow for testing of the code's functionality for figure generation. While viral sequences have been deposited in the GISAID database as part of the Danish SARS-CoV-2 surveillance response (EPI_SET_260124ub for the time period of interest), a list of specific accession numbers is not provided in this manuscript. This is because the linkage of specific genomic identifiers to the sensitive clinical and demographic metadata analyzed here could potentially allow for the re-identification of individuals, which is prohibited under Danish law.

Software overview

R (4.2.3), Python (3.10.10), brms (2.20.3), Guppy, Medaka, iVar (1.4.3), BCFtools (1.18), VADR, Trim Galore (0.6.10).

Authors

• Mark Khurana (mark.khurana@sund.ku.dk)
• Alexandros Katsiferis (alexandros.katsiferis@sund.ku.dk)
• Neil Scheidwasser (neil.clow@sund.ku.dk)

License

Apache 2.0 License

Citation

To be added.