Single cell references (Brain)

This example used microarray transcriptomic profiling of 41 patients from different age groups GSE53890.

As references a single cell dataset from 466 cells (GSE67835) published by Darmanis et al., 2015 (Proc Natl Acad Sci).

import anndata
import seaborn as sns
import matplotlib.pyplot as plt
import dtangle

# Load dataset
adata_mix = anndata.read_h5ad("../../examples/data/brain_aging_microarray.h5ad")
adata_ref = anndata.read_h5ad("../../examples/data/brain_single_cell.h5ad")
adata_mix.var_names_make_unique()
adata_ref.var_names_make_unique()

/home/malte/Dokumente/Github/dtangle/.venv/lib/python3.11/site-packages/anndata/_core/anndata.py:1825: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")

Run deconvolution from single cell reference

dtangle.deconvolut(
    adata_mix,
    references=adata_ref,
    reference_annotation_col="cell_type",
    data_type="microarray-gene",
)

adata_mix.obsm["dtangle"].head()

	Oligodendrocyte precursor cells	Neurons	Neuroblast	Astrocytes	Oligodendrocytes	Microglia
GSM1303144_24M.CEL	0.130750	0.385471	0.157636	0.128135	0.087965	0.110044
GSM1303145_25BF.CEL	0.134050	0.373632	0.158939	0.135272	0.088278	0.109829
GSM1303146_25F.CEL	0.131120	0.372604	0.154251	0.146045	0.091005	0.104975
GSM1303147_26BM.CEL	0.129604	0.356444	0.152652	0.157790	0.091801	0.111709
GSM1303148_26M.CEL	0.129315	0.357495	0.153784	0.141635	0.101902	0.115869

proportions = adata_mix.obsm["dtangle"].copy()
proportions["age_group"] = adata_mix.obs["age_group"].values
proportions = proportions.groupby("age_group", observed=True).mean()
proportions.plot(kind="bar", stacked=True, figsize=(6, 6), width=0.8)
plt.ylabel("Estimated Cell Type Proportion")
plt.legend(bbox_to_anchor=(1.05, 1), loc="upper left", frameon=False)
plt.tight_layout()
plt.show()

fig, ax = plt.subplots(figsize=(6, 6))
sns.violinplot(adata_mix.obsm["dtangle"], ax=ax)
sns.stripplot(adata_mix.obsm["dtangle"], color="k", size=3, ax=ax)
ax.set_xticks(ax.get_xticks())
_ = ax.set_xticklabels(ax.get_xticklabels(), rotation=45, ha="right")

merge_df = adata_mix.obsm["dtangle"].merge(adata_mix.obs["age_group"], left_index=True, right_index=True)
fig, ax = plt.subplots(figsize=(6, 6))
sns.violinplot(
    x="age_group",
    y="Neurons",
    data=merge_df,
    order=[
        "young",
        "middle aged",
        "normal aged",
        "extremely aged",
    ],
    inner="quart",
    hue="age_group",
    ax=ax,
    palette="Set2",
)
sns.stripplot(
    x="age_group",
    y="Neurons",
    data=merge_df,
    order=[
        "young",
        "middle aged",
        "normal aged",
        "extremely aged",
    ],
    color="k",
    ax=ax,
)
ax.set_xlabel("Age Group")
ax.set_ylabel("Estimated Neuron Proportion")

Text(0, 0.5, 'Estimated Neuron Proportion')