Loading [MathJax]/extensions/Safe.js

######## snakemake preamble start (automatically inserted, do not edit) ########
import sys;sys.path.extend(['/fh/fast/bloom_j/software/miniforge3/envs/dms-vep-pipeline-3_v3.17/lib/python3.11/site-packages', '/fh/fast/bloom_j/computational_notebooks/jbloom/2024/RABV_Pasteur_G_DMS/dms-vep-pipeline-3/..', '/fh/fast/bloom_j/computational_notebooks/jbloom/2024/RABV_Pasteur_G_DMS/dms-vep-pipeline-3', '/fh/fast/bloom_j/software/miniforge3/envs/dms-vep-pipeline-3_v3.17/bin', '/fh/fast/bloom_j/software/miniforge3/envs/dms-vep-pipeline-3_v3.17/lib/python3.11', '/fh/fast/bloom_j/software/miniforge3/envs/dms-vep-pipeline-3_v3.17/lib/python3.11/lib-dynload', '/home/jbloom/.local/lib/python3.11/site-packages', '/fh/fast/bloom_j/software/miniforge3/envs/dms-vep-pipeline-3_v3.17/lib/python3.11/site-packages', '/home/jbloom/.cache/snakemake/snakemake/source-cache/runtime-cache/tmp0j08q3w9/file/fh/fast/bloom_j/computational_notebooks/jbloom/2024/RABV_Pasteur_G_DMS/notebooks', '/fh/fast/bloom_j/computational_notebooks/jbloom/2024/RABV_Pasteur_G_DMS/notebooks']);import pickle;from snakemake import script;script.snakemake = pickle.loads(b'\x80\x04\x95L\x10\x00\x00\x00\x00\x00\x00\x8c\x10snakemake.script\x94\x8c\tSnakemake\x94\x93\x94)\x81\x94}\x94(\x8c\x05input\x94\x8c\x0csnakemake.io\x94\x8c\nInputFiles\x94\x93\x94)\x81\x94(\x8cGresults/summaries/all_antibodies_and_cell_entry_per_antibody_escape.csv\x94\x8c3results/summaries/all_antibodies_and_cell_entry.csv\x94e}\x94(\x8c\x06_names\x94}\x94(\x8c\x17per_antibody_escape_csv\x94K\x00N\x86\x94\x8c\x0ephenotypes_csv\x94K\x01N\x86\x94u\x8c\x12_allowed_overrides\x94]\x94(\x8c\x05index\x94\x8c\x04sort\x94eh\x15h\x06\x8c\x0eAttributeGuard\x94\x93\x94)\x81\x94}\x94\x8c\x04name\x94h\x15sbh\x16h\x18)\x81\x94}\x94h\x1bh\x16sbh\x0fh\nh\x11h\x0bub\x8c\x06output\x94h\x06\x8c\x0bOutputFiles\x94\x93\x94)\x81\x94(\x8c*results/dms-viz/domainIII_CR57_escape.json\x94\x8c9results/dms-viz/domainIII_CR57_escape_no_description.json\x94\x8c1results/dms-viz/domainIII_CR57_escape_sitemap.csv\x94\x8c4results/dms-viz/domainIII_CR57_escape_phenotypes.csv\x94\x8c4results/dms-viz/domainIII_CR57_escape_description.md\x94\x8c)results/dms-viz/domainIII_CR57_escape.pdb\x94e}\x94(h\r}\x94(\x8c\x04json\x94K\x00N\x86\x94\x8c\x13json_no_description\x94K\x01N\x86\x94\x8c\x07sitemap\x94K\x02N\x86\x94\x8c\nphenotypes\x94K\x03N\x86\x94\x8c\x0edescription_md\x94K\x04N\x86\x94\x8c\x08pdb_file\x94K\x05N\x86\x94uh\x13]\x94(h\x15h\x16eh\x15h\x18)\x81\x94}\x94h\x1bh\x15sbh\x16h\x18)\x81\x94}\x94h\x1bh\x16sbh*h"h,h#h.h$h0h%h2h&h4h\'ub\x8c\x06params\x94h\x06\x8c\x06Params\x94\x93\x94)\x81\x94(}\x94(\x8c\x06pdb_id\x94\x8c\x048r40\x94\x8c\x13escape_or_phenotype\x94\x8c\x06escape\x94\x8c\nphenotypes\x94N\x8c\nantibodies\x94]\x94\x8c\x04CR57\x94a\x8c\x0fincluded-chains\x94]\x94\x8c\x01C\x94a\x8c\x0fexcluded-chains\x94]\x94\x8c\x08alphabet\x94\x8c\x14RKHDEQNSTYWFAILMVGPC\x94\x8c\x05floor\x94\x88\x8c\x0csummary-stat\x94\x8c\x03sum\x94\x8c\x0ctooltip-cols\x94}\x94\x8c\ncell entry\x94\x8c\ncell entry\x94s\x8c\x0bfilter-cols\x94}\x94\x8c\ncell entry\x94\x8c\ncell entry\x94s\x8c\rfilter-limits\x94}\x94\x8c\ncell entry\x94]\x94(J\xf8\xff\xff\xffJ\xfb\xff\xff\xffK\x00es\x8c\x0eheatmap-limits\x94]\x94K\x00a\x8c\x05title\x94\x8c=Effects of mutations to rabies G on escape from antibody CR57\x94\x8c\x0bdescription\x94\x8c\x84Pseudovirus deep mutational scanning of how mutations to rabies G affect\nneutralization by antibody CR57 shown on domain III (8r40)\n\x94u\x8c\xcaStudy by [Aditham et al](https://www.biorxiv.org/content/10.1101/2024.12.17.628970v1) (2024).\nSee [https://github.com/dms-vep/RABV_Pasteur_G_DMS](https://github.com/dms-vep/RABV_Pasteur_G_DMS) for code.\x94e}\x94(h\r}\x94(\x8c\x06config\x94K\x00N\x86\x94\x8c\x12description_suffix\x94K\x01N\x86\x94uh\x13]\x94(h\x15h\x16eh\x15h\x18)\x81\x94}\x94h\x1bh\x15sbh\x16h\x18)\x81\x94}\x94h\x1bh\x16sbhgh?hihdub\x8c\twildcards\x94h\x06\x8c\tWildcards\x94\x93\x94)\x81\x94\x8c\x15domainIII_CR57_escape\x94a}\x94(h\r}\x94\x8c\x06struct\x94K\x00N\x86\x94sh\x13]\x94(h\x15h\x16eh\x15h\x18)\x81\x94}\x94h\x1bh\x15sbh\x16h\x18)\x81\x94}\x94h\x1bh\x16sb\x8c\x06struct\x94htub\x8c\x07threads\x94K\x01\x8c\tresources\x94h\x06\x8c\tResources\x94\x93\x94)\x81\x94(K\x01K\x01\x8c\x14/loc/scratch/3868073\x94e}\x94(h\r}\x94(\x8c\x06_cores\x94K\x00N\x86\x94\x8c\x06_nodes\x94K\x01N\x86\x94\x8c\x06tmpdir\x94K\x02N\x86\x94uh\x13]\x94(h\x15h\x16eh\x15h\x18)\x81\x94}\x94h\x1bh\x15sbh\x16h\x18)\x81\x94}\x94h\x1bh\x16sbh\x87K\x01h\x89K\x01h\x8bh\x84ub\x8c\x03log\x94h\x06\x8c\x03Log\x94\x93\x94)\x81\x94\x8c?results/notebooks/configure_dms_viz_domainIII_CR57_escape.ipynb\x94a}\x94(h\r}\x94\x8c\x08notebook\x94K\x00N\x86\x94sh\x13]\x94(h\x15h\x16eh\x15h\x18)\x81\x94}\x94h\x1bh\x15sbh\x16h\x18)\x81\x94}\x94h\x1bh\x16sbh\x99h\x96ubhg}\x94(\x8c\rpipeline_path\x94\x8c\x12dms-vep-pipeline-3\x94\x8c\x04docs\x94\x8c\x04docs\x94\x8c\x08homepage\x94\x8c\x0fhomepage/public\x94\x8c\x18build_vitepress_homepage\x94\x88\x8c\x0fgithub_repo_url\x94\x8c-https://github.com/dms-vep/RABV_Pasteur_G_DMS\x94\x8c\x0fgithub_blob_url\x94\x8c7https://github.com/dms-vep/RABV_Pasteur_G_DMS/blob/main\x94\x8c\x0bdescription\x94\x8c5Deep mutational scanning of rabies G (Pasteur strain)\x94\x8c\x04year\x94M\xe8\x07\x8c\x07authors\x94\x8cL[Aditham et al](https://www.biorxiv.org/content/10.1101/2024.12.17.628970v1)\x94\x8c\x1euse_precomputed_barcode_counts\x94\x89\x8c\x12site_numbering_map\x94\x8c\x1bdata/site_numbering_map.csv\x94\x8c\x1emutation_design_classification\x94}\x94(\x8c\x03csv\x94\x8c\x1bdata/designed_mutations.csv\x94\x8c\x08site_col\x94\x8c\x0fsequential_site\x94u\x8c\x16neut_standard_barcodes\x94\x8c)data/neutralization_standard_barcodes.csv\x94\x8c\x11prebuilt_variants\x94N\x8c\x10prebuilt_geneseq\x94N\x8c\x0bpacbio_runs\x94\x8c\x14data/PacBio_runs.csv\x94\x8c\x0fpacbio_amplicon\x94\x8c\x17data/PacBio_amplicon.gb\x94\x8c\x15pacbio_amplicon_specs\x94\x8c$data/PacBio_feature_parse_specs.yaml\x94\x8c\x0cvariant_tags\x94}\x94(\x8c\x0cvariant_tag5\x94}\x94(\x8c\tvariant_1\x94\x8c\x01G\x94\x8c\tvariant_2\x94hJ\x8c\x08wildtype\x94\x8c\x01A\x94u\x8c\x0cvariant_tag3\x94}\x94(\x8c\tvariant_1\x94h\xc9\x8c\tvariant_2\x94hJ\x8c\x08wildtype\x94h\xccuu\x8c\x12max_ccs_error_rate\x94G?\x1a6\xe2\xeb\x1cC-\x8c\x10consensus_params\x94}\x94(\x8c\rmax_sub_diffs\x94N\x8c\x0fmax_indel_diffs\x94N\x8c\x12max_minor_sub_frac\x94G?\xc9\x99\x99\x99\x99\x99\x9a\x8c\x14max_minor_indel_frac\x94G?\xc9\x99\x99\x99\x99\x99\x9a\x8c\x0bmin_support\x94K\x03u\x8c\x13gene_sequence_codon\x94\x8c\x1edata/gene_sequence/codon.fasta\x94\x8c\x15gene_sequence_protein\x94\x8c data/gene_sequence/protein.fasta\x94\x8c\x0ecodon_variants\x94\x8c#results/variants/codon_variants.csv\x94\x8c\x0cbarcode_runs\x94\x8c\x15data/barcode_runs.csv\x94\x8c\x12duplicate_fastq_R1\x94\x8c\x04warn\x94\x8c\x1eillumina_barcode_parser_params\x94}\x94(\x8c\x08upstream\x94\x8c"ACTCCACTAGGAACATTTCTCTCTCGAATCTAGA\x94\x8c\ndownstream\x94\x8c\x00\x94\x8c\x04minq\x94K\x14\x8c\x11upstream_mismatch\x94K\x02u\x8c\x13func_effects_config\x94\x8c\x1cdata/func_effects_config.yml\x94\x8c\x16antibody_escape_config\x94\x8c\x1fdata/antibody_escape_config.yml\x94\x8c\x10summaries_config\x94\x8c\x19data/summaries_config.yml\x94\x8c\x0edms_viz_config\x94\x8c\x17data/dms_viz_config.yml\x94u\x8c\x04rule\x94\x8c\x11configure_dms_viz\x94\x8c\x0fbench_iteration\x94N\x8c\tscriptdir\x94\x8cQ/fh/fast/bloom_j/computational_notebooks/jbloom/2024/RABV_Pasteur_G_DMS/notebooks\x94ub.');del script;from snakemake.logging import logger;from snakemake.script import snakemake; logger.printshellcmds = False;import os; os.chdir(r'/fh/fast/bloom_j/computational_notebooks/jbloom/2024/RABV_Pasteur_G_DMS');
######## snakemake preamble end #########

Configure dms-viz JSONs

Imports:

import gzip
import os
import requests
import subprocess
import warnings

import Bio.PDB.PDBParser
import Bio.PDB.Polypeptide

import matplotlib.colors

import pandas as pd

import seaborn

Get variables from snakemake:

config = snakemake.params.config
pdb_id = config["pdb_id"]
escape_or_phenotype = config["escape_or_phenotype"]
phenotypes_list = config["phenotypes"]
antibody_list = config["antibodies"]

phenotypes_csv = snakemake.input.phenotypes_csv
if escape_or_phenotype == "escape":
    per_antibody_escape_csv = snakemake.input.per_antibody_escape_csv

dms_viz_json = snakemake.output.json
dms_viz_json_no_description = snakemake.output.json_no_description
dms_viz_sitemap = snakemake.output.sitemap
dms_viz_phenotypes = snakemake.output.phenotypes
pdb_file = snakemake.output.pdb_file
description_md = snakemake.output.description_md

description_suffix = snakemake.params.description_suffix

name = snakemake.wildcards.struct

Build the sitemap used by dms-viz:

phenotypes_all = pd.read_csv(phenotypes_csv)

sitemap = (
    phenotypes_all
    [["sequential_site", "site", "wildtype"]]
    .rename(columns={"site": "reference_site"})
    .drop_duplicates()
    .sort_values("sequential_site")
)

assert len(sitemap) == sitemap["sequential_site"].nunique() == sitemap["reference_site"].nunique()

sitemap.to_csv(dms_viz_sitemap, index=False)

Get the biological assembly (see https://pdb101.rcsb.org/learn/guide-to-understanding-pdb-data/biological-assemblies#Anchor-download) in case the crystallographic unit doesn't correspond to that:

pdb_id = config["pdb_id"]

print(f"Getting PDB file for {pdb_id}")

r = requests.get(f"https://files.rcsb.org/download/{pdb_id}.pdb1.gz")
assert r.status_code == 200
pdb_content = gzip.decompress(r.content).decode("utf-8")
with open(pdb_file, "w") as f:
    f.write(pdb_content)

Getting PDB file for 8r40

Check the sites mismatched between the sitemap and the protein structure in terms of residue identity:

with warnings.catch_warnings():
    warnings.simplefilter("ignore")
    pdb_obj = Bio.PDB.PDBParser().get_structure(id=pdb_id, file=pdb_file)[0]

all_chains = [chain.id for chain in pdb_obj]
print(f"PDB has {all_chains=}")
if "included-chains" in config:
    included_chains = config["included-chains"]
    print(f"{included_chains=}")
    assert set(included_chains).issubset(all_chains)
else:
    included_chains = all_chains

records = []
for chain in included_chains:
    for res in pdb_obj[chain].get_residues():
        if not res.id[0].isspace():
            continue
        aa = Bio.PDB.Polypeptide.protein_letters_3to1[res.resname]
        r = res.id[1]
        records.append((chain, r, aa))
pdb_df = pd.DataFrame(records, columns=["chain", "reference_site", "pdb_aa"])

mismatched_sites = sitemap.merge(pdb_df, how="left")

print(
    f"Of {len(sitemap)} sites, {len(mismatched_sites.query('wildtype == pdb_aa'))} match, "
    f"{len(mismatched_sites.query('pdb_aa.isnull()'))} are missing from PDB, and "
    f"{len(mismatched_sites.query('pdb_aa.notnull()').query('wildtype != pdb_aa'))} differ."
)

print("Sites that differ:")
display(mismatched_sites.query("pdb_aa.notnull() and (wildtype != pdb_aa)").reset_index(drop=True))

PDB has all_chains=['R', 'A', 'G', 'C']
included_chains=['C']
Of 433 sites, 81 match, 344 are missing from PDB, and 8 differ.
Sites that differ:

	sequential_site	reference_site	wildtype	chain	pdb_aa
0	42	40	G	C	E
1	189	187	M	C	T
2	190	188	S	C	P
3	206	204	S	C	N
4	207	205	E	C	K
5	226	224	K	C	R
6	249	247	N	C	D
7	257	255	G	C	D

Process the different phenotypes and escape if relevant:

# get extra columns dealing with fact `dms-viz` does not allow them to have spaces
extra_cols = []
for coltype in ["tooltip-cols", "filter-cols"]:
    for col in list(config[coltype]):
        col_nospace = col.replace(" ", "_")
        assert (col in phenotypes_all.columns) or (col_nospace in phenotypes_all.columns), f"no column {col} in {phenotypes_all.columns=}"
        if col != col_nospace:
            assert (col_nospace not in phenotypes_all.columns) or (col not in phenotypes_all.columns)
            assert col_nospace not in config[coltype]
            phenotypes_all = phenotypes_all.rename(columns={col: col_nospace}, errors="ignore")
            config[coltype][col_nospace] = config[coltype][col]
            del config[coltype][col]
            extra_cols.append(col_nospace)
            if coltype == "filter-cols":
                if col in config["filter-limits"]:
                    config["filter-limits"][col_nospace] = config["filter-limits"][col]
                    del config["filter-limits"][col]
        else:
            extra_cols.append(col)
print(f"Keeping {extra_cols=} for filters and tooltips")
assert set(extra_cols).issubset(phenotypes_all.columns)

id_cols = ["site", "wildtype", "mutant"]

if escape_or_phenotype == "phenotype":
    assert phenotypes_list and not antibody_list, f"{phenotypes_list=}, {antibody_list=}"
    assert set(phenotypes_list).issubset(phenotypes_all.columns), f"{phenotypes_list=}\n{phenotypes_all.columns=}"
    metric_column = "effect"
    condition_column = "phenotype"
    phenotypes = (
        phenotypes_all
        .melt(
            id_vars=id_cols + extra_cols,
            value_vars=phenotypes_list,
            var_name=condition_column,
            value_name=metric_column,
        )
    )

elif escape_or_phenotype == "escape":
    assert not phenotypes_list

    phenotypes = pd.read_csv(per_antibody_escape_csv)

    antibodies_all = list(phenotypes["antibody"].unique())
    if antibody_list is not None:
        antibodies = antibody_list
        print(f"Using just {antibodies=} from {antibodies_all=}")
        assert set(antibodies).issubset(antibodies_all), f"{antibodies=}\n{antibodies_all=}"
        assert len(antibodies) >= 1
    else:
        antibodies = antibodies_all
        print(f"Using all {antibodies=}")
    antibodies = antibodies
    metric_column = "escape"
    condition_column = "antibody"
    phenotypes = phenotypes[id_cols + [metric_column, condition_column]]
    phenotypes = phenotypes[phenotypes[condition_column].isin(antibodies)]
    
    if extra_cols:
        join_data = phenotypes_all[id_cols + extra_cols]
        assert len(join_data) == len(join_data.groupby(id_cols))
        assert set(join_data.columns).intersection(phenotypes.columns) == set(id_cols)
        phenotypes = phenotypes.merge(join_data, on=id_cols, validate="many_to_one", how="left")

else:
    raise ValueError(f"invalid {escape_or_phenotype=}")

print(f"Writing the phenotypes to {dms_viz_phenotypes}")
phenotypes.to_csv(dms_viz_phenotypes, index=False)
display(phenotypes)

Keeping extra_cols=['cell_entry', 'cell_entry'] for filters and tooltips

Using just antibodies=['CR57'] from antibodies_all=['17C7', 'CR4098', 'CR57', 'CTB012', 'RVA122', 'RVC20', 'RVC58', 'RVC68']

Writing the phenotypes to results/dms-viz/domainIII_CR57_escape_phenotypes.csv

	site	wildtype	mutant	escape	antibody	cell_entry	cell_entry
0	-2	F	A	-0.014410	CR57	-0.44860	-0.44860
1	-2	F	C	0.027760	CR57	-0.18730	-0.18730
2	-2	F	D	-0.012160	CR57	0.01876	0.01876
3	-2	F	E	-0.030880	CR57	-0.79690	-0.79690
4	-2	F	G	-0.074570	CR57	-0.02989	-0.02989
...	...	...	...	...	...	...	...
6242	431	D	S	-0.024780	CR57	-0.26610	-0.26610
6243	431	D	T	-0.021390	CR57	-0.17420	-0.17420
6244	431	D	V	-0.012870	CR57	-0.76310	-0.76310
6245	431	D	W	0.029020	CR57	-0.56050	-0.56050
6246	431	D	Y	-0.008859	CR57	-0.72370	-0.72370

6247 rows × 7 columns

Get enough colors:

def get_hex_color_palette(num_colors):
    colors = seaborn.color_palette("hls", num_colors)
    hex_colors = [matplotlib.colors.to_hex(color) for color in colors]
    return hex_colors

nconditions = phenotypes[condition_column].nunique()
if nconditions > 4:
    colors = ",".join(get_hex_color_palette(nconditions))
else:
    colors = "#0072B2,#CC79A7,#4C3549,#009E73"

print(f"Using {colors=}")
seaborn.palplot(seaborn.color_palette(colors.split(",")))

Using colors='#0072B2,#CC79A7,#4C3549,#009E73'

Get the description:

description_text = "\n\n".join(
    [
        f"## {config['title']}",
        config["description"] if config["description"] is not None else "",
        f"Structure shown is from PDB {pdb_id}.",
        description_suffix,
    ]
).lstrip()

print(description_text)

with open(description_md, "w") as f:
    f.write(description_text)

## Effects of mutations to rabies G on escape from antibody CR57

Pseudovirus deep mutational scanning of how mutations to rabies G affect
neutralization by antibody CR57 shown on domain III (8r40)


Structure shown is from PDB 8r40.

Study by [Aditham et al](https://www.biorxiv.org/content/10.1101/2024.12.17.628970v1) (2024).
See [https://github.com/dms-vep/RABV_Pasteur_G_DMS](https://github.com/dms-vep/RABV_Pasteur_G_DMS) for code.

Now run configure-dms-viz:

cmds = [
    "configure-dms-viz", "format",
    "--name", name,
    "--input", dms_viz_phenotypes,
    "--output", dms_viz_json_no_description,
    "--structure", pdb_file,
    "--metric", metric_column,
    "--condition", condition_column,
    "--sitemap", dms_viz_sitemap,
    "--colors", colors,
]

for col in [
    "included-chains",
    "excluded-chains",
]:
    val = config[col]
    if val:
        cmds += [f"--{col}", " ".join(val)]

if config["heatmap-limits"]:
    cmds += ["--heatmap-limits", ", ".join(str(x) for x in config["heatmap-limits"])]

for col in [
    "alphabet",
    "floor",
    "summary-stat",
    "tooltip-cols",
    "filter-cols",
    "filter-limits",
    "title",
    "description",
]:
    cmds += [f"--{col}", str(config[col])]

print(f"Running the following commands:\n{cmds}")
subprocess.run(cmds, check=True)

cmds_join = [
    "configure-dms-viz", "join",
    "--input", dms_viz_json_no_description,
    "--description", description_md,
    "--output", dms_viz_json,
]

print(f"Running the following commands:\n{cmds_join}")
subprocess.run(cmds_join, check=True)

Running the following commands:
['configure-dms-viz', 'format', '--name', 'domainIII_CR57_escape', '--input', 'results/dms-viz/domainIII_CR57_escape_phenotypes.csv', '--output', 'results/dms-viz/domainIII_CR57_escape_no_description.json', '--structure', 'results/dms-viz/domainIII_CR57_escape.pdb', '--metric', 'escape', '--condition', 'antibody', '--sitemap', 'results/dms-viz/domainIII_CR57_escape_sitemap.csv', '--colors', '#0072B2,#CC79A7,#4C3549,#009E73', '--included-chains', 'C', '--heatmap-limits', '0', '--alphabet', 'RKHDEQNSTYWFAILMVGPC', '--floor', 'True', '--summary-stat', 'sum', '--tooltip-cols', "{'cell_entry': 'cell entry'}", '--filter-cols', "{'cell_entry': 'cell entry'}", '--filter-limits', "{'cell_entry': [-8, -5, 0]}", '--title', 'Effects of mutations to rabies G on escape from antibody CR57', '--description', 'Pseudovirus deep mutational scanning of how mutations to rabies G affect\nneutralization by antibody CR57 shown on domain III (8r40)\n']

Formatting data for visualization using the 'escape' column from 'results/dms-viz/domainIII_CR57_escape_phenotypes.csv'...

Using sitemap from 'results/dms-viz/domainIII_CR57_escape_sitemap.csv'.

'protein_site' column is not present in the sitemap. Assuming that the reference sites correspond to protein sites.

One value was provided for heatmap limits, this will be the center value of the scale.

Warning: The 'cell_entry' filter limit '-8' is less than the minimum value of -7.526. Setting the min value to -7.526.

About 90.91% (80 of 88) of the wildtype residues in the data match the corresponding residues in the structure.
Warning: 79.58% (343 of 431) of the data sites are missing from the structure.

Success! The visualization JSON was written to 'results/dms-viz/domainIII_CR57_escape_no_description.json'

Running the following commands:
['configure-dms-viz', 'join', '--input', 'results/dms-viz/domainIII_CR57_escape_no_description.json', '--description', 'results/dms-viz/domainIII_CR57_escape_description.md', '--output', 'results/dms-viz/domainIII_CR57_escape.json']

Success! 1 JSON files were merged and saved to 'results/dms-viz/domainIII_CR57_escape.json'

CompletedProcess(args=['configure-dms-viz', 'join', '--input', 'results/dms-viz/domainIII_CR57_escape_no_description.json', '--description', 'results/dms-viz/domainIII_CR57_escape_description.md', '--output', 'results/dms-viz/domainIII_CR57_escape.json'], returncode=0)