feat: analysis of low cfDNA input updated for comparison to QG

This commit is contained in:
2025-10-17 09:30:11 +02:00
parent 2581de3dc7
commit 7a35a49084
3 changed files with 227 additions and 43 deletions

View File

@@ -1,6 +1,6 @@
import marimo
__generated_with = "0.14.16"
__generated_with = "0.16.4"
app = marimo.App(width="medium")
@@ -168,7 +168,7 @@ def _(mo):
def _(os, pd, qc_hist_dir):
# QG historical data
qg_hist_stats = pd.read_csv(
os.path.join(qc_hist_dir, "historical_data_LB.csv"), index_col=0
os.path.join(qc_hist_dir, "historical_data_LB-QG.csv"), index_col=0
)
qg_hist_stats.columns = [i.replace("|", " / ") for i in qg_hist_stats.columns]
qg_hist_stats = qg_hist_stats.drop(["batch"], axis=1)
@@ -866,6 +866,33 @@ def _(clinicals, meta, pd, stats):
return (stats_meta,)
@app.cell
def _(pd):
meta_qg = pd.read_excel(
"/home/darren/Documents/4_data/3_internal/2_lb/QC/lb2_metadata.xlsx",
index_col="fastq_id",
)
meta_qg
return (meta_qg,)
@app.cell
def _(meta_qg, pd, qg_hist_stats):
# merge stats and metadata and remove seracares
stats_meta_qg = pd.concat([qg_hist_stats, meta_qg], axis=1)
stats_meta_qg = stats_meta_qg.loc[
[
i
for i in stats_meta_qg.index
if "twist" not in i.lower() and "sera" not in i.lower()
],
:,
]
stats_meta_qg = stats_meta_qg.reset_index()
stats_meta_qg.head()
return (stats_meta_qg,)
@app.cell
def _(mo, stats_meta):
var1 = mo.ui.dropdown(options=stats_meta.columns, label="Variable 1")
@@ -975,6 +1002,108 @@ def _(stats_meta):
return
@app.cell
def _(stats_meta_melt):
stats_meta_melt["Flow cell #"].unique()
return
@app.cell
def _(qg_hist_stats):
qg_hist_stats
return
@app.cell
def _(pd, stats_meta, stats_meta_qg):
stats_meta["lab"] = "Serenomica"
stats_meta_qg["lab"] = "Quantgene"
all_stats_meta = pd.concat([stats_meta, stats_meta_qg], axis=0)
all_stats_meta.shape
return (all_stats_meta,)
@app.cell
def _(all_stats_meta):
all_stats_meta["ng_group"] = [
"< 30ng" if i < 30 else "> 30ng"
for i in all_stats_meta["Input DNA used (ng)"]
]
all_stats_meta[["lab", "ng_group"]].value_counts()
return
@app.cell
def _(all_stats_meta, pd):
all_stats_meta_melt = pd.melt(
all_stats_meta,
id_vars=["index", "sample_type", "ng_group", "Flow cell #", "lab"],
)
all_stats_meta_melt
return (all_stats_meta_melt,)
@app.cell
def _(all_stats_meta_melt, mo):
box_vars2 = mo.ui.dropdown(
options=all_stats_meta_melt["variable"].unique(), label="Variables to plot"
)
mo.hstack([box_vars2])
return (box_vars2,)
@app.cell
def _(all_stats_meta_melt, box_vars2, px):
fig7 = px.box(
data_frame=all_stats_meta_melt.loc[
(all_stats_meta_melt["variable"] == box_vars2.value)
],
color="ng_group",
y="value",
x="lab",
template="simple_white",
points="all",
labels={
"ng_group": "cfDNA Input for Library",
"value": box_vars2.value,
"lab": "Laboratory",
},
color_discrete_sequence=px.colors.qualitative.Dark2,
category_orders={"ng_group": ["< 30ng", "> 30ng"]},
hover_data=["index"],
width=800,
)
fig7.update_traces(
marker=dict(size=10, line=dict(width=1, color="DarkSlateGrey")),
selector=dict(type="points"),
)
fig7.update_xaxes(tickfont_size=14)
fig7.update_yaxes(showgrid=True, tickfont_size=14)
fig7.update_legends(font_size=14)
for trace2 in fig7.select_traces():
trace2.marker.update(
size=7, line=dict(width=1, color="DarkSlateGrey"), opacity=0.8
)
fig7.show()
return
@app.cell
def _(all_stats_meta):
all_stats_meta[["lab", "ng_group"]].value_counts()
return
@app.cell
def _(stats_meta):
stats_meta.sort_values("Input DNA used (ng)")[
["index", "Input DNA used (ng)", "Flow cell #"]
]
return
@app.cell
def _():
return