Spatial Metabolite Analysis

%load_ext autoreload
%autoreload
import scanpy as sc
import spmetatme.plotting as pl
from spmetatme.utils import get_metabolite_adata

import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)
warnings.filterwarnings("ignore")

c:\Users\suraj\anaconda3\envs\spatialEnv\Lib\site-packages\tqdm\auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
  from .autonotebook import tqdm as notebook_tqdm

Load Spatial metabolic fluxes

Structure of the loaded AnnData object:

• ``.X``: Reaction fluxes matrix (rows = spatial spots, columns = reactions)

• ``.obs``: Metadata for each spot (metabolic domains)

• ``.var``: Metadata for each reaction (‘rxn_id’, ‘rxn_names’, ‘rxn_full_names’, ‘ex_rxns’, ‘subsystems’)

• ``.obsm[‘metabolites’]``: The key data! A matrix of net metabolite balance (rows = spots, columns = metabolites)

• ``.uns``: Unstructured metadata including ‘TME_metabolites’, ‘met_names’, ‘moranI’, ‘neighbors’, ‘spatial’

adata = sc.read_h5ad(r"output/metabolic_Breast_cancer_Block_A.h5ad")
adata

AnnData object with n_obs × n_vars = 3798 × 6561
    obs: 'domain'
    var: 'rxn_id', 'rxn_names', 'rxn_full_names', 'ex_rxns', 'subsystems'
    uns: 'TME_metabolites', 'met_names', 'moranI', 'neighbors', 'spatial'
    obsm: 'ex_rxns', 'metabolites', 'spatial'
    obsp: 'communication', 'connectivities', 'distances'

From spatial metabolic adata inferred by spMetaTME, get metabolite level adata object

• n_obs: Number of spatial spots (same as before)

• n_vars: Number of metabolites

• ``.X``: net metabolite balance (spots × metabolites)

• ``.obs``: spot metadata (domain)

• ``.var``: metabolite information:

  • metabolite_names: Common name (e.g., “glucose”, “L-lactate”)

  • Compartment: Where metabolite is located (code: c, e, m, p)

  • Compartment_name: Full compartment name (e.g., “Cytoplasm”, “Extracellular”)

  • moran_score: Spatial autocorrelation metric (high = spatially clustered, low = randomly distributed)

metabolite_adata = get_metabolite_adata(adata)
metabolite_adata

AnnData object with n_obs × n_vars = 3798 × 2919
    obs: 'domain'
    var: 'Compartment', 'Compartment_name', 'metabolite_names'
    uns: 'TME_metabolites', 'met_names', 'moranI', 'neighbors', 'spatial'
    obsm: 'ex_rxns', 'metabolites', 'spatial'
    obsp: 'distances', 'connectivities'

Spatial Maps of Metabolites

Parameters:

• metabolite_names: List of metabolite names in human-readable format

• metabolites: List of metabolite model IDs (alternative way to specify metabolites)

• img_key: Which image to use as background (default: “hires” for high-resolution image)

• size: Size of spots in the plot (adjust for visibility)

• figsize_per_row: Dimensions of figure per metabolite row

• file_name (optional): filename to save the plot

pl.plot_spatial_metabolites(metabolite_adata, metabolite_names=['glucose','pyruvate','L-lactate'])

Plot spatial metabolite distribution by metabolite id

pl.plot_spatial_metabolites(metabolite_adata, metabolites=['MAM02403c','MAM02403e'])

Metabolite distribution across compartments and domains

pl.metabolite_ridges_plot(metabolite_adata, n_cols=3)

Identify differentially enriched metabolites

df = pl.plot_differential_metabolite_heatmap(metabolite_adata, top_n=5, return_marker_df=True)
df

	group	names	scores	logfoldchanges	pvals	pvals_adj
0	0	ornithine [c]	43.213764	0.265625	0.000000e+00	0.000000e+00
1	0	palmitolate [c]	43.115639	0.349242	0.000000e+00	0.000000e+00
133	0	NADP+ [m]	30.568985	0.787096	3.163521e-205	2.798278e-204
302	0	sphinganine [c]	23.213366	-0.841104	3.337094e-119	1.245649e-118
385	0	tyrosine [e]	20.521523	-1.777718	1.383148e-93	4.103057e-93
...	...	...	...	...	...	...
14445	4	NADP+ [m]	-33.108196	-0.761299	2.265194e-240	2.755042e-239
14488	4	palmitate [r]	-33.682312	-0.325301	1.049550e-248	1.730868e-247
14591	4	sphinganine [c]	-42.286964	1.775674	0.000000e+00	0.000000e+00
14592	4	1-phosphatidyl-1D-myo-inositol-4-phosphate [c]	-42.633022	2.909879	0.000000e+00	0.000000e+00
14593	4	cys-gly [e]	-42.761959	-1.398427	0.000000e+00	0.000000e+00

115 rows × 6 columns