def fill_sample_report(page_n, sample_store):
page_n = int(page_n)
sample_ids = list(map(lambda x: x["_id"], sample_store))
if len(sample_ids) == 0:
return None
data_table = import_data.filter_all(sample_ids=sample_ids,
pagination={
"page_size": SAMPLE_PAGESIZE,
"current_page": page_n
})
max_page = len(sample_store) // SAMPLE_PAGESIZE
# We need to have fake radio buttons with the same ids to account for times
# when not all SAMPLE_PAGESIZE samples are shown and are not taking the ids required by the callback
html_fake_radio_buttons = html.Div([
dcc.RadioItems(options=[{
'label': '',
'value': 'nosample'
}],
value='noaction',
id="sample-radio-{}".format(n_sample))
for n_sample in range(len(data_table), SAMPLE_PAGESIZE)
],
style={"display": "none"})
return [
html.H4("Page {} of {}".format(page_n + 1, max_page + 1)),
html.Div(children_sample_list_report(data_table)),
html_fake_radio_buttons,
admin.html_qc_expert_form(),
html.H4("Page {} of {}".format(page_n + 1, max_page + 1)),
dcc.ConfirmDialog(
id='qc-confirm',
message='Are you sure you want to send sample feedback?',
)
]
def update_selected_samples(n_clicks, param_store, collection_name, run_names,
species_list, species_source, group_list, qc_list,
sample_names, prev_sample_store, date_seq_start,
date_seq_end):
date_range = [date_seq_start, date_seq_end]
for i in range(2):
if date_range[i] is not None:
date_range[i] = datetime.datetime.strptime(
re.split('T| ', date_range[i])[0], '%Y-%m-%d')
if sample_names is not None and sample_names != "":
sample_names = sample_names.split("\n")
else:
sample_names = param_store.get("sample_names", [])
if not run_names:
run_names = param_store.get("run", [])
if not group_list:
group_list = param_store.get("group", [])
if not species_list:
species_list = param_store.get("species", [])
if not qc_list:
qc_list = param_store.get("qc", [])
if not date_range[0]:
date_range[0] = param_store.get("date_seq_start", None)
if not date_range[1]:
date_range[1] = param_store.get("date_seq_end", None)
#override if selected collection
if collection_name is not None:
run_names = [collection_name]
if (date_range[0] is None and date_range[1] is None):
date_range = None
if (n_clicks == 0 and sample_names == [] and run_names == []
and group_list == [] and species_list == [] and qc_list == []
and date_range is None):
samples = prev_sample_store
else:
samples = import_data.filter_all(species=species_list,
species_source=species_source,
group=group_list,
qc_list=qc_list,
run_names=run_names,
sample_names=sample_names,
date_range=date_range,
projection={"name": 1})
if "_id" in samples:
samples["_id"] = samples["_id"].astype(str)
samples = samples.to_dict('records')
# if deleted_samples:
# samples = [s for s in samples if s["_id"] not in deleted_samples]
return samples
def generate_download_button(download_button, run_names, species_list,
species_source, group_list, qc_list,
sample_names):
if download_button == 0:
return None
else:
if sample_names is not None and sample_names != "":
sample_names = sample_names.split("\n")
tests_df = import_data.filter_all(species=species_list,
species_source=species_source,
group=group_list,
qc_list=qc_list,
run_names=run_names,
sample_names=sample_names,
pagination=None)
# return samples.to_dict()
if not len(tests_df):
return None
tests_df = generate_table(tests_df)
rename_dict = {item["id"]: item["name"] for item in global_vars.COLUMNS}
renamed = tests_df.rename(rename_dict, axis='columns')
missing_columns = [
a for a in list(rename_dict.values()) if not a in list(renamed.columns)
]
# add missing columns
for column in missing_columns:
renamed[column] = np.nan
# reorder columns
renamed = renamed[list(rename_dict.values())]
csv_string_eur = renamed.to_csv(index=False,
encoding="utf-8",
sep=";",
decimal=",")
tsv_string_us = renamed.to_csv(index=False, encoding="utf-8", sep="\t")
full_csv_string_eur = 'data:text/csv;charset=utf-8,' + \
urlparse.quote(csv_string_eur)
full_tsv_string_us = 'data:text/tab-separated-values;charset=utf-8,' + \
urlparse.quote(tsv_string_us)
return [
html.A("(tsv, US format)",
href=full_tsv_string_us,
download='report.tsv'), " - ",
html.A("(csv, EUR Excel format)",
href=full_csv_string_eur,
download='report.csv')
]
def update_filter_table(_, sample_store):
if len(sample_store) == 0:
return ["0", [{}], False]
sample_ids = list(map(lambda x: x["_id"], sample_store))
samples = import_data.filter_all(sample_ids=sample_ids)
samples = generate_table(samples)
if len(sample_store) > 500:
virtualization = True
else:
virtualization = False
return [len(sample_store), samples.to_dict("rows"), virtualization]
def pipeline_report_data(sample_data):
if len(sample_data) == 0:
return [], [{
"id": "Priority",
"name": "Priority"
}, {
"id": "Sample",
"name": "Sample"
}, {
"id": "QC status",
"name": "QC status"
}], [], [], []
status_dict = {
"Success": "OK",
"Running": "Running",
"initialized": "init.",
"Failure": "Fail",
"Requirements not met": "Req.",
"queued to run": "queue",
}
samples = import_data.filter_all(
sample_ids=[s["_id"] for s in sample_data],
projection={
"properties.stamper": 1,
'name': 1,
'sample_sheet.priority': 1,
"components": 1
})
s_c_components = []
for _id, sample in samples.iterrows():
try:
for component in sample["components"]:
s_c_components.append(component["name"])
except TypeError:
pass
components_list = [
comp for comp in components_order if comp in s_c_components
]
rows = []
columns = [{
"name": "Priority",
"id": "priority"
}, {
"name": "Sample",
"id": "sample"
}, {
"name": "QC status",
"id": "qc_val"
}]
rerun_form_components = []
for comp in components_list:
columns.append({"name": comp, "id": comp})
# HERE
rerun_form_components.append({"label": comp, "value": comp})
# Conditional data colors
style_data_conditional = [{
"if": {
"column_id": "qc_val",
"filter_query": '{qc_val} contains "CF"'
},
"backgroundColor": "#ea6153"
}, {
"if": {
"column_id": "qc_val",
"filter_query": '{qc_val} contains "CF(LF)"'
},
"backgroundColor": "#ea6153"
}, {
"if": {
"column_id": "qc_val",
"filter_query": '{qc_val} contains "OK"'
},
"backgroundColor": "#27ae60"
}, {
"if": {
"column_id": "qc_val",
"filter_query": '{qc_val} eq "SL"'
},
"backgroundColor": "#f1c40f"
}]
for col in components_list:
style_data_conditional.append({
"if": {
"column_id": col,
"filter_query": '{{{}}} eq "Fail"'.format(col)
},
"backgroundColor": "#ea6153"
})
style_data_conditional.append({
"if": {
"column_id": col,
"filter_query": '{{{}}} eq "OK"'.format(col)
},
"backgroundColor": "#3498db"
})
style_data_conditional.append({
"if": {
"column_id": col,
"filter_query": '{{{}}} eq "Running"'.format(col)
},
"backgroundColor": "#f1c40f"
})
style_data_conditional.append({
"if": {
"column_id": col,
"filter_query": '{{{}}} eq "Req."'.format(col)
},
"backgroundColor": "#d3d3d3",
"color": "#525252"
})
rerun_form_samples = []
for _id, sample in samples.iterrows():
name = sample["name"]
row = {}
if name == "Undetermined":
continue # ignore this row
row["sample"] = name
row["_id"] = str(sample["_id"])
rerun_form_samples.append({
"label": name,
"value": "{}:{}".format(_id, name)
})
priority = str(sample.get("sample_sheet.priority", "")).lower()
# prio_display = " " Emoji not supported
# if priority == "high":
# prio_display = "High"
# else:
# prio_display = "Low"
row["priority"] = priority
qc_val = sample.get("properties.stamper.summary.stamp.value", "N/A")
if pd.isna(qc_val):
qc_val = "N/A"
expert_check = False
expert_stamp = sample.get('stamps.supplying_lab_check.value')
if expert_stamp is not None and not pd.isna(expert_stamp):
qc_val = sample.get('stamps.supplying_lab_check.value')
expert_check = True
statusname = ""
if qc_val == "supplying lab":
qc_val = "SL"
statusname = "status-1"
elif qc_val == "N/A":
statusname = "status--2"
elif (qc_val == "core facility"):
statusname = "status--1"
qc_val = "CF"
elif qc_val == "OK":
statusname = "status-2"
qc_val = "OK"
if expert_check:
try:
qc_val += "*"
except:
print(sample)
row["qc_val"] = qc_val
try:
for component in sample["components"]:
row[component["name"]] = status_dict[component["status"]]
row["{}_id".format(component["name"])] = str(component["_id"])
except TypeError:
pass
except KeyError:
print(sample["_id"])
print(sample["components"])
rows.append(row)
def sort_name(e):
return e["sample"]
rows.sort(key=sort_name)
return rows, columns, style_data_conditional, rerun_form_components, rerun_form_samples
def link_to_files(sample_store):
"""Generates a script string """
sample_ids = [str(s["_id"]) for s in sample_store]
samples = import_data.filter_all(sample_ids=sample_ids,
projection={"name": 1,
"properties.datafiles.summary.paired_reads": 1})
samples["_id"] = samples["_id"].astype(str)
samples = samples.set_index("_id")
# Access samples by their id
assemblies = import_data.get_assemblies_paths(sample_ids)
reads_script = "mkdir samples\ncd samples\n"
assemblies_script = "mkdir assemblies\ncd assemblies\n"
reads_errors = []
assemblies_errors = []
samples_total = set()
for _id, sample in samples.iterrows():
try:
reads_script += "#{}\nln -s {} .\nln -s {} .\n".format(
sample["name"],
sample["properties.datafiles.summary.paired_reads"][0],
sample["properties.datafiles.summary.paired_reads"][1])
samples_total.add((sample["name"], sample.name))
except (KeyError, TypeError):
reads_errors.append("Missing data for sample: {} - {}. In database:\n{}".format(
sample.get("name", "None"),
sample.name,
sample.get(
"properties.datafiles.summary.paired_reads", "No data")
))
if len(reads_errors):
reads_html = [
html.Div([
html.Div([
html.H6("Read files",
className="m-0 font-weight-bold text-primary"),
], className="card-header py-3"),
html.Div([
"A few errors occurred locating the read paths. If you need more info, " +
"please contact an admin.",
html.Pre("\n".join(reads_errors), className="error-pre"),
html.Pre(reads_script, className="folder-pre")
], className="card-body")
], className="card shadow")
]
else:
reads_html = [
html.Div([
html.Div([
html.H6("Read files",
className="m-0 font-weight-bold text-primary"),
], className="card-header py-3"),
html.Div([
html.Pre(reads_script, className="folder-pre")
], className="card-body")
], className="card shadow")
]
samples_assemblies = set()
for assembly in assemblies:
sample = samples.loc[str(assembly["sample"]["_id"]), :]
assemblies_script += "#{}\nln -s {} {}\n".format(
sample["name"],
assembly["path"] + "/contigs.fasta",
sample["name"] + "_contigs.fasta")
samples_assemblies.add((sample["name"], sample.name))
assemblies_missing = samples_total - samples_assemblies
for sample in assemblies_missing:
assemblies_errors.append("Missing data for sample: {} - {}.".format(sample[0], sample[1]))
if len(assemblies_errors):
assemblies_html = [
html.Div([
html.Div([
html.H6("Assemblies",
className="m-0 font-weight-bold text-primary"),
], className="card-header py-3"),
html.Div([
"A few errors occurred locating the contigs. If you need more info, " +
"please contact an admin.",
html.Pre("\n".join(assemblies_errors),
className="error-pre"),
html.Pre(assemblies_script, className="folder-pre")
], className="card-body")
], className="card mb-4 shadow")
]
else:
assemblies_html = [
html.Div([
html.Div([
html.H6("Assemblies",
className="m-0 font-weight-bold text-primary"),
], className="card-header py-3"),
html.Div([
html.Pre(assemblies_script, className="folder-pre")
], className="card-body")
], className="card mb-4 shadow")
]
return html.Div([html.Div(assemblies_html), html.Div(reads_html)])