This repository is used to provide a molecular epidemiology analysis of SARS-CoV-2 data from GISAID.

• Installation
• Repo contents
• Usage
• License
• Issues

The molecular-epi-ncov package requires a standard server or computer with enough RAM to provide support for operations performed in memory.

All code is tested on Linux: Ubuntu 20.04.1 and MacOSX operating systems.

The R script is compatible with Windows, Mac, and Linux operating systems.

You will require the following software installed on your server or computer before starting:

• Python(>=3.9)
• Jupyter Notebook
• Treetime
• Nextclade
• Augur
• Seqkit
• R(>=4.2)

From a terminal, use pip to install the following Python dependencies before running the script:

pip install pandas seaborn matplotlib

From an R session, type:

install.packages("reshape2", "ggplot2", "htmlwidgets", "webshot")
devtools::install_github("hrbrmstr/streamgraph")

• ancestral_sequence: Mutation table of ancestral sequences reconstructed by treetime.
• clade-distribution: R package code.
• exampledata: Example data randomly extracted from the Gisaid database and China: Guangdong sequences.
• accession_number.txt: Accession numbers of China: Guangdong sequences.
• participant_characteristics.png: SARS-CoV-2 seroprevalence survey and participant characteristics

To obtain the most recent GISAID SARS-CoV-2 data in a single file, you can use the batch download feature on the GISAID website（https://gisaid.org/）. The files you need are metadata_tsv_2024_01_27.tar.xz and sequences_fasta_2024_01_27.tar.xz. Please note that membership is required to access this feature.

You can download the SARS-CoV-2 reference dataset, to run :

nextclade dataset get \
  --name 'nextstrain/sars-cov-2/wuhan-hu-1' \
  --tag '2024-01-28T00:00:00Z' \
  --output-dir '~/sars-cov-2-2024-01-28update'

To filter sequences that meet the criteria suspiciously clustered single-nucleotide polymorphisms (SNPs) [quality control (QC) SNP clusters status metric not “good”; ≥ 6 mutations in 100 bases], too many private mutations (QC private mutations status metric not good; ≥ 10 mutations from the nearest tree node), or overall bad quality (Nextclade QC overall status “bad”), run seq_filter.ipynb(global sequences) or gd_seq.ipynb(China: Guangdong sequences) in Jupyter Notebook or VS Code.

Run clade_distribution.ipynb to obtain weekly_clade_distribution.csv, then run clade-distribution/scripts/clade.R to obtain genotype distribution plots.

To get the mutation heatmap of SARS-CoV-2, run mutation_heatmap.ipynb. Ancestral sequences were reconstructed using TreeTime. Taking ba.5.2.48 as an example, run the script sh ancestral_sequence.sh to retrieve its ancestral node. The command treetime ancestral is computationally intensive and requires a significant amount of memory.

This project is covered under MIT License.