Studying the mutational fitness effects of SARS-CoV-2 spike fusion peptide

Dependencies

• Python (version 3.9)
• Biopython
• Seaborn
• Matplotlib
• R (version 4.1)
• PEAR
• PyMOL

Input files

• Raw read files in fastq format from NIH SRA database BioProject PRJNA910585
• ./Fasta/SARS2-FP_with_flank.fa: Fusion peptide sequences with 21 nt upstream (5' flank)
• ./Fasta/FP_ref.fa: Reference (i.e. wild type) amino acid sequences (primer regions not included)
• ./data/Tree_aa_fitness.csv: S mutational fitness estimated from phylogeny by Bloom & Neher. The orignal data are from here.
• ./data/BA1_DMS_muteffects_observed.csv: S mutational fitness measured by a pseudovirus system by Dadonaite et al. The orignal data are from here.

Primer design for DMS library construction

1. Generating foward (NNK + internal barcode) and reverse primers (constant)
   python3 script/lib_primer_design.py

   • Input file:
     • ./Fasta/SARS2-FP_with_flank.fa
   • Output files:
     • ./primer/SARS2-FP_lib_Fprimer_bc.fa
     • ./primer/SARS2-FP_lib_Rprimer_bc.fa

2. Generating barcode file
   python3 script/check_barcode.py

   • Input files:
     • ./primer/SARS2-FP_lib_Fprimer_bc.fa
     • ./Fasta/SARS2-FP_with_flank.fa
   • Output file:
     • ./data/barcodes.tsv

Calculating fitness from DMS data

1. Merge overlapping paired-end reads using PEAR
   pear -f [FASTQ FILE FOR FORWARD READ] -r [FASTQ FILE FOR FORWARD READ] -o [OUTPUT FASTQ FILE]

   • Output files should be placed in the fastq_merged/ folder and named as described in ./doc/filename_merged_fastq.tsv

2. Counting variants based on nucleotide sequences
   python3 script/FP_fastq2count.py

   • Input files:
     • Merged read files in fastq_merged/ folder
   • Output files:
     • result/FP_DMS_count_nuc.tsv

3. Convert nucleotide sequences to amino acid mutations
   python3 script/FP_count_nuc2aa.py

   • Input files:
     • ./data/barcodes.tsv
     • ./Fasta/FP_ref.fa
     • result/FP_DMS_count_nuc.tsv
   • Output files:
     • ./result/FP_DMS_count_aa.tsv

4. Convert nucleotide sequences to codon variants
   python3 script/FP_count_nuc2codon.py

   • Input files:
     • ./data/barcodes.tsv
     • ./Fasta/FP_ref.fa
     • result/FP_DMS_count_nuc.tsv
   • Output files:
     • ./result/FP_DMS_count_codon.tsv

5. Compute fitness
   python3 script/FP_count2fit.py

   • Input files:
     • ./result/FP_DMS_count_aa.tsv
   • Output file:
     • ./result/FP_DMS_fit.tsv
     • ./result/FP_DMS_fit_by_resi.tsv

6. Convert B factor in PDB file into mean fitness value
   python3 script/convert_Bfactor_to_fit.py

   • Input file:
     • ./PDB/7my8.pdb
     • ./result/FP_DMS_fit_by_resi.tsv
   • Output file:
     • ./PDB/7my8_fit.pdb

Plotting

1. Plot correlation between replicates and compare silent/missense/nonsense
   Rscript script/plot_QC.R

   • Input file:
     • ./result/FP_DMS_fit.tsv
   • Output files:
     • graph/QC_*.png

2. Plot correlation between fitness measurements in this study and those in previous studies python3 script/plot_cor_measures.py

   • Input file:
     • ./result/FP_DMS_fit.tsv
     • ./data/BA1_DMS_muteffects_observed.csv
     • ./data/Tree_aa_fitness.csv
   • Output files:
     • graph/cor_*.png

3. Plot heatmap for the fitnss of individual mutations
   Rscript script/plot_heatmap_fit.R

   • Input file:
     • ./result/FP_DMS_fit.tsv
   • Ouput file:
     • graph/FP_fit_heatmap_*.png

4. Plot mean fitness (i.e. mutational tolerance) of individual residue positions
   Rscript script/plot_mean_fit.R

   • Input file:
     • ./result/FP_DMS_fit_by_resi.tsv
   • Output file:
     • ./graph/mean_fit.png

5. Plot meanfitness on structure
   pymol script/plot_Bfactor_as_fit.pml

   • Input file:
     • ./PDB/7my8_fit.pdb
   • Output file:
     • ./graph/structure_FP.png

6. Plot heatmap for the antibody escape of individual mutations
   Rscript script/plot_heatmap_escape.R

   • Input file:
     • ./result/FP_DMS_fit.tsv
   • Ouput file:
     • graph/FP_escape_*.png

7. Plot heatmap for the codon variants
   Rscript script/plot_heatmap_codon_freq.R

   • Input file:
     • ./result/FP_DMS_codon_freq.tsv
   • Ouput file:
     • graph/FP_codon_freq_*.png

Analyze mutation rate

1. Identity the number of amino acid mutations on each merged read
   python3 script/analyze_mut_rate.py

   • Input file:
     • Merged read files in fastq_merged/ folder
     • ./Fasta/FP_ref.fa
   • Output file:
     • ./result/Lib1_mut_count.tsv
     • ./result/Lib2_mut_count.tsv

2. Plot mutation rate
   Rscript script/plot_lib_mut_count.R

   • Input file:
     • ./result/Lib1_mut_count.tsv
     • ./result/Lib2_mut_count.tsv
   • Output file:
     • ./graph/Mut_count_input_lib.png