Skip to content

Latest commit

 

History

History
53 lines (34 loc) · 7.99 KB

README.md

File metadata and controls

53 lines (34 loc) · 7.99 KB
Raw

Input data

This directory contains input data for the analysis.

Basic information about sequences and alignments

These files are used for the basic processing of the deep sequencing data to call variants by barcode and count barcodes:

  • PacBio_amplicons.gb: the amplicons being sequenced by PacBio. Note that there are a variety of possible amplicons because in addition to the SARS-CoV-2 RBD there are RBDs from a variety of other viral strains.

  • feature_parse_specs.yaml: how to parse the amplicon when handling the PacBio data.

  • PacBio_runs.csv: list of the PacBio runs used to call the variants.

  • barcode_runs.csv: list of the Illumina runs used to count the barcodes for different samples. This file is built from barcode_runs_orig-names.csv by the Jupyter notebook build_barcode_runs.ipynb.

  • RBD_sites.csv: gives site and residue information for SARS-CoV-2, including alignment of the RBD integer numbering with the Spike numbering for SARS-CoV-2 RBD, alignment to SARS-CoV, and structural annotations as detailed below.

  • ./plasmid_maps: gives our base SARS-CoV-2 yeast display vector sequence, including the modifications made for a barcode landing pad per our library generation scheme. 2649 illustrates what the plasmid looks like after the insertion of a mutagenized amplicon with an appended N16 barcode.

  • ./primers/ contains important primer sequences used in the study, including the design of mutagenesis primer pools, and additional primers used in mutant library construction, amplicon barcoding, and Illumina sequencing preparation.

Alignments of different Spikes / RBDs

The ./alignments/ subdirectory contains alignments and phylogenetic trees of Spike and RBD constructs from human, bat, and other mammal sarbecovirus isolates. Sequences were aligned both as nucleotide gene sequecnes and as translated amino acid sequences. The script ./alignments/alignment-and-phylogeny.sh was used to align sequences with mafft and infer phylogenetic trees with raxml. The unaligned sequence inputs to this script are found in ./alignments/unaligned-sequences/, with each fasta header providing common names for sarbecovirus isolates and their genome accession number in NCBI or GISAID. Sequence selection includes:

  • The curated set of 30 sequences from Letko, Marzi, and Munster (2020), which contained all known unique RBD sequences at the time of its publication. See Extended Data Fig 1 for more information.
  • Bat sarbecovirus isolate WIV16, which was removed from the Letko dataset because it has an identical RBD amino acid sequence to WIV1, and isolate HKU3-1, which was removed from the Letko dataset because it has an identical RBD amino acid sequence to HKU3-13.
  • RaTG13 and RmYN02, two newly described bat CoV sequences that are the most closely related strains currently known to SARS-CoV-2 (at least at the whole-genome level).
  • Two recent pangolin CoV sequences from Lam et al. (2020), including the infectious virus isolated from a pangolin seized in Guanxi, and the consensus sequence from two closely related isolates from pangolins seized in Guangdong. The consensus sequence was constructed from an alignment of the GD pangolin CoV in GISAID accssion EPI_ISL_410544, and the reconstructed metagenomic viral genome reported in the Lam et al. (2020) supplement.
  • BtKY72, a bat coronavirus isolated in Kenya which, along with the Bulgaria sample BM48-31 is used to root the phylogeny

The ./alignments/Spike_GISAID/ directory contains the .zip file of Spike files downloaded from GISAID, along with a script to modify header names to remove spaces and align sequences using mafft. The output of this script is used as input to the circulating_variants analysis. We acknowledge all GISAID contributors, listed in ./alignments/Spike_GISAID/gisaid_hcov-19_acknowledgement_table.xls

Input data from previous studies

The ./lit-measurements subdirectory contains raw data on sarbecovirus functional properties from previous experimental studies, which we will in comparison to our deep mutational scanning measurements. Files include:

Isogenic validations

The ./isogenic_titrations/ subdirectory contains processed data from single-clone experiments to test out the yeast display system and validate mutational effects from our deep mutational scanning experiments. Details are provided in .md summaries from each .Rmd analysis conducted within this subdirectory.

Structures

The ./structures/ directory contains information relevant for structural analyses. Specifically, ./structures/Ab-bound/ and ./structures/ACE2-bound contain PDB files of SARS-CoV and SARS-CoV-2 spike and RBD proteins, either free, bound to ACE2, or antibodies. Each subdirectory contains the .pdb files, and a .pse session illustrating structural alignment of the component .pdb files.

The ./structures/annotate-structural-elements.R script analyzes the various PDB files to create strucutral annotations, including relative solvent accessibility, ACE2 contact residues, and antibody epitope sites. This script was run to create the file ./RBD_sites.csv

The file ./structures/surface_constraint_features.pse was constructed from a set of pdb structures in this directory along with the output pdb files from the structure_function.Rmd script, which replaces the RBD b factor column with the mean mutational effects per site from our DMS measurements. By executing the list of commands in ./structures/surface_constraint_commands.txt in PyMol, this .pse can be generated anew. This PyMol session contains structural alignments of ACE2-bound RBD, full Spike trimer in the closed state, and all mAb structures included in our analyses. The RBD is loaded as two different objects, where the surface is colored by the mean mutational effect on binding or expression (colored from red to white, where any site with mean mutational effect values less than or equal to -2 is the darkest red, and mean effects from -2 to zero scale from red to white).

The images in ./structures/epitopes_view2.pdf were generated from this PyMol session by following the commands and instructions given iin ./structures/images_epitope_commands.txt.

Fonts