Zika virus E protein deep mutational scanning

Deep mutational scanning of Zika virus E protein. Experiments performed by the Matt Evans lab. Sequencing and computational analyses performed by Danny Lawrence and Jesse Bloom in the Bloom lab.

Quick summary

Look at the Jupyter notebook analysis_notebook.ipynb, or its markdown output at results/summary/analysis_notebook.md for the results of the deep mutational scanning. The growth curve data are analyzed and plotted by plot_growth_data.ipynb, and its markdown output is at results/summary/plot_growth_data.md.

Running the notebooks

You can run the notebooks with the run_nbs.bash:

bash run_nbs.bash

Or to submit it to the server:

sbatch -c 16 -p largenode --mem=150000 run_nbs.bash

Analysis and results

The analysis is performed by the Jupyter notebook analysis_notebook.ipynb using dms_tools2. That notebook also contains plots and descriptions of the results.

The results files are placed in the ./results subdirectory. Most of the results are not tracked in this GitHub repo, but some are. Specifically:

• ./results/codoncounts/ contains files that give the counts of each codon mutation for each sample from the barcoded subamplicon sequencing.

• ./results/prefs/ has the amino-acid preferences for each library, as well as the across-library un-scaled preferences (./results/prefs/unscaled_prefs.csv) and the re-scaled preferences (./results/prefs/rescaled_prefs.csv). For most purposes, this last file is the one you want.

• ./results/muteffects has the mutational effects calculated from the amino-acid preferences.

• ./results/diffsel/ has the differential selection for each library, as well as the across-library average mutation differential selection (./results/diffsel/summary_ZKA64-meanmutdiffsel.csv for ZKA185 and ./results/diffsel/summary_ZKA185-meanmutdiffsel.csv for ZKA185).

• ./results/fracsurvive/ has the excess (above-average) fraction surviving for each library, as well as the across-library excess fraction surviving (./results/diffsel/summary_ZKA64-meanmutfracsurvive.csv for ZKA185 and ./results/diffsel/summary_ZKA185-meanmutfracsurvive.csv for ZKA185).

• ./results/logoplots contains logo plots visualizing the amino-acid preferences, differential selection, and excess fraction surviving.

• ./results/figures contains plots that we anticipate being figures in the paper.

Input data

The input data are in the ./data/ subdirectory. These data consist of:

• ./data/E.fasta: coding sequence of E protein from ZIKV MR766 strain.

• ./data/samplelist.csv: all the samples that we sequenced and the locations of the associated deep-sequencing data. Although we have technical replicates for the antibody selections for each library, we group these and just analyze variation at the level of biological replicates.

• ./data/subamplicon_alignspecs.txt: the alignment specs for the barcoded subamplicon sequencing.

• ./data/E_alignment.fasta: alignment of ZIKV E protein created by Danny Lawrence. Downloaded from NCBI Virus Variation Resource all ZIKV sequences (2019-03-26) with the query settings as follows:

  • Sequence type: Protein
  • Host: any
  • Region/Country: any
  • Genome region: E
  • Isolation source: any
  • Collapse identical sequences: checked ("Note: All groups of identical sequences in the dataset will be represented by the oldest sequence in the group.")
  • Added the E region of our WT sequence for MR766 as the reference sequence and used phydms_prepalignment to trim the alignment and remove any sequences that are redundant or incomplete. A number of sequences appeared to be derived from an experiment or otherwise spurious and were removed manually.

• data/domains.csv: the domain structure of E as defined in Figure 1 of Dai et al (2016).

• ./data/all_growth_data.csv: Growth curve data generated by Matt Evans.

• PDB structures of the E protein:

  • data/5ire.pdb: PDB 5ire

  • data/5ire_monomer.pdb: PDB 5ire, chain A only

  • data/5u4w.pdb: PDB 5u4w

  • data/6co8.pdb: PDB 5co8

  • data/6co8_monomer.pdb: PDB 5co8, chain A only

• Secondary structure and solvent accessibility calculations made by running the dssp webserver on the above structures:

  • data/5ire.dssp

  • data/5ire_monomer.dssp

  • data/6co8.dssp

  • data/6co8_monomer.dssp

• Plasmid maps:

  • data/1725_ZIKV_MR766_int_WT.gb: plasmid for wild-type ZIKV reverse genetics.

  • data/1726_ZIKV_MR766_int_GFP.gb: recipient plasmid for cloning mutant libraries.

PyMol scripts

The subdirectory pymol_scripts contains scripts for generating PyMol structure images.