[1]:

from genecover import *
import numpy as np
import scanpy as sc
import pandas as pd
import os

Marker Gene Selection from DLPFC Sample #151673

Load Dataset

[3]:

data_dir = "..\\data\\DLPFC"
file_folder = "151673"
path = os.path.join(data_dir, file_folder)
adata = sc.read_visium(path,count_file = "filtered_feature_bc_matrix.h5",load_images=True)
adata.var_names_make_unique()
df_meta = pd.read_csv(os.path.join(path, "metadata.tsv"), sep='\t')
df_meta_layer = df_meta['layer_guess']
adata.obs['ground_truth'] = df_meta_layer.values
adata = adata[~pd.isnull(adata.obs['ground_truth'])]
sc.pp.filter_genes(adata, min_cells=100)
sc.pp.normalize_total(adata, target_sum=1e4)
sc.pp.log1p(adata)
orig_gene = adata.var.index.values

C:\Users\An Wang\AppData\Local\Temp\ipykernel_60420\1400016560.py:4: FutureWarning: Use `squidpy.read.visium` instead.
  adata = sc.read_visium(path,count_file = "filtered_feature_bc_matrix.h5",load_images=True)
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\scanpy\preprocessing\_simple.py:287: ImplicitModificationWarning: Trying to modify attribute `.var` of view, initializing view as actual.
  adata.var["n_cells"] = number

Computing Correlation Matrix

[4]:

corr_mat = gene_gene_correlation(adata.X.toarray())

GeneCover via Combinatorial Optimization (Gurobi Solver)

[5]:

# Obtain 100 marker genes
genecover_markers = GeneCover(num_marker=100, corr_mat=corr_mat, w=np.ones(corr_mat.shape[1]), solver = "Gurobi")
print("GeneCover markers: \n", orig_gene[genecover_markers])

Set parameter Username
Academic license - for non-commercial use only - expires 2026-04-07
Best Gap:  0
Best Epsilon:  0.1311767578125
GeneCover markers:
 ['MARCKSL1' 'SYNC' 'ATP1A1' 'ATP1B1' 'NME7' 'CNTN2' 'HPCAL1' 'VSNL1'
 'MDH1' 'PPP3R1' 'CTNNA2' 'TMSB10' 'IGKC' 'MAL' 'ERMN' 'LRP2' 'GAD1'
 'CHN1' 'MOBP' 'CCK' 'AC106707.1' 'CLDN11' 'FAM131A' 'LDB2' 'UCHL1' 'SPP1'
 'BBS7' 'HHIP' 'TMEM144' 'GPM6A' 'SLC1A3' 'ENC1' 'EDIL3' 'CXCL14' 'GABRB2'
 'MOG' 'ACTB' 'NDUFA4' 'RAPGEF5' 'AQP1' 'PHKG1' 'YWHAG' 'SLC26A4-AS1'
 'GJB1' 'NAP1L2' 'NEFM' 'NEFL' 'STMN2' 'CALB1' 'ENPP2' 'DIRAS2' 'CERCAM'
 'OLFM1' 'PTGDS' 'SAA1' 'FOLH1' 'MYRF' 'SCGB2A2' 'CARNS1' 'NRGN' 'NKX6-2'
 'NELL2' 'KRT8' 'KCNC2' 'SYT1' 'CUX2' 'CABP1' 'AACS' 'RTN1' 'HSPA2'
 'PPP4R4' 'GLDN' 'HBA2' 'SLC5A11' 'AC009133.1' 'PLLP' 'NDRG4' 'CALB2'
 'KRT19' 'CNP' 'GFAP' 'NSF' 'ANKRD40' 'AATK' 'MBP' 'CHGB' 'SNAP25' 'CST3'
 'BCAS1' 'NDUFA7' 'PPP1R14A' 'CALM3' 'RSPH14' 'NEFH' 'YWHAH' 'C21orf91'
 'OLIG1' 'PCP4' 'MT-CO1' 'MT-CO2']

GeneCover via Combinaotorial Optimization (SCIP Solver)

[ ]:

genecover_markers = GeneCover(num_marker=100, corr_mat=corr_mat, w=np.ones(corr_mat.shape[1]), solver = "SCIP")
print("GeneCover markers via SCIP: \n", orig_gene[genecover_markers])

Best Gap:  0
Best Epsilon:  0.12958984375
GeneCover markers via SCIP:
 ['PADI2' 'SYNC' 'WDR47' 'ATP1A1' 'ATP1B1' 'NME7' 'CNTN2' 'VSNL1' 'PREPL'
 'MDH1' 'PPP3R1' 'TMSB10' 'IGKC' 'MAL' 'ERMN' 'LRP2' 'GAD1' 'CHN1' 'CCK'
 'CLDN11' 'LDB2' 'SPP1' 'HHIP' 'TMEM144' 'GPM6A' 'SLC1A3' 'ENC1' 'EDIL3'
 'RHOBTB3' 'CXCL14' 'GABRB2' 'MOG' 'TPBG' 'ACTB' 'NDUFA4' 'RAPGEF5' 'AQP1'
 'YWHAG' 'SLC26A4-AS1' 'CD99' 'GJB1' 'NAP1L2' 'LAMP2' 'NEFM' 'NEFL'
 'STMN2' 'CALB1' 'ENPP2' 'CERCAM' 'OLFM1' 'PTGDS' 'SAA1' 'FOLH1' 'MYRF'
 'FTH1' 'SCGB2A2' 'PLA2G16' 'RTN3' 'CARNS1' 'HSPA8' 'NRGN' 'NKX6-2'
 'TUBA1B' 'KRT8' 'KCNC2' 'SYT1' 'CUX2' 'CABP1' 'AACS' 'RTN1' 'HSPA2'
 'PPP4R4' 'GLDN' 'NPTN' 'HBA1' 'CRYM' 'SLC5A11' 'AC009133.1' 'PLLP'
 'NDRG4' 'CALB2' 'KRT19' 'KRT17' 'GFAP' 'NSF' 'ANKRD40' 'AATK' 'MBP'
 'SNAP25' 'CST3' 'MAG' 'PPP1R14A' 'CALM3' 'RSPH14' 'NEFH' 'YWHAH' 'OLIG1'
 'PCP4' 'MT-CO1' 'MT-CO2']

GeneCover via Greedy Heuristics

[8]:

#obtain 100 marker genes via greedy heuristics
genecover_markers_greedy = GeneCover(num_marker=100, corr_mat=corr_mat, w=np.ones(corr_mat.shape[1]), solver = "Greedy")
print("GeneCover markers: \n", orig_gene[genecover_markers_greedy])

Best Gap:  0
Best Epsilon:  0.136181640625
GeneCover markers:
 ['SNAP25' 'ERMN' 'ENC1' 'VSNL1' 'MAG' 'SCGB2A2' 'ATP1B1' 'RTN1' 'PTGDS'
 'OLFM1' 'MOG' 'MBP' 'FABP4' 'IGKC' 'CST3' 'CARNS1' 'GAD1' 'TMEM144'
 'KRT19' 'MT-CO1' 'NME7' 'UCHL1' 'PPP1R14A' 'CHN1' 'CLDN11' 'HBB' 'HSPA2'
 'CALB2' 'EDIL3' 'YWHAG' 'GJB1' 'NEFL' 'ENPP2' 'KRT8' 'RNASE1' 'SLC5A11'
 'GFAP' 'YWHAH' 'SYNC' 'RPL5' 'ATP1A1' 'HPCAL1' 'FAM84A' 'PPP3R1' 'TMSB10'
 'MOBP' 'LDB2' 'LIMCH1' 'SPP1' 'HHIP' 'SV2C' 'SLC12A2' 'CXCL14' 'TUBB2A'
 'ACTB' 'RAPGEF5' 'AQP1' 'PHKG1' 'SEMA3E' 'CHRDL1' 'LAMP2' 'NEFM' 'CALB1'
 'DIRAS2' 'SLC44A1' 'STXBP1' 'CERCAM' 'ABCA2' 'FOLH1' 'FTH1' 'THY1'
 'HSPA8' 'NRGN' 'NKX6-2' 'VAMP1' 'SYT1' 'CABP1' 'AACS' 'HTR2A' 'SLAIN1'
 'HS6ST3' 'GLDN' 'IQCK' 'GPRC5B' 'AC009133.1' 'PLLP' 'NDRG4' 'RPL26'
 'LRRC75A' 'KRT17' 'NSF' 'ANKRD40' 'SEPT4' 'AATK' 'CHGB' 'LAMP5' 'NEFH'
 'OLIG1' 'S100B' 'MT-CO3']

Iterative GeneCover via Combinatorial Optimization (Gurobi Solver)

[9]:

# Obtain 200 marker genes via Iterative GeneCover with an incremental size of 100 and two iterations
genecover_markers_iterative = Iterative_GeneCover(incremental_sizes=[100,100], corr_mat=corr_mat, w=np.ones(corr_mat.shape[1]), solver = "Gurobi")

Iteration 1
Best Gap:  0
Best Epsilon:  0.1311767578125
Iteration  2
Best Gap:  0
Best Epsilon:  0.1367919921875

[10]:

print("Iterative GeneCover markers: \n", orig_gene[genecover_markers_iterative])

Iterative GeneCover markers:
 [['MARCKSL1' 'SYNC' 'ATP1A1' 'ATP1B1' 'NME7' 'CNTN2' 'HPCAL1' 'VSNL1'
  'MDH1' 'PPP3R1' 'CTNNA2' 'TMSB10' 'IGKC' 'MAL' 'ERMN' 'LRP2' 'GAD1'
  'CHN1' 'MOBP' 'CCK' 'AC106707.1' 'CLDN11' 'FAM131A' 'LDB2' 'UCHL1'
  'SPP1' 'BBS7' 'HHIP' 'TMEM144' 'GPM6A' 'SLC1A3' 'ENC1' 'EDIL3' 'CXCL14'
  'GABRB2' 'MOG' 'ACTB' 'NDUFA4' 'RAPGEF5' 'AQP1' 'PHKG1' 'YWHAG'
  'SLC26A4-AS1' 'GJB1' 'NAP1L2' 'NEFM' 'NEFL' 'STMN2' 'CALB1' 'ENPP2'
  'DIRAS2' 'CERCAM' 'OLFM1' 'PTGDS' 'SAA1' 'FOLH1' 'MYRF' 'SCGB2A2'
  'CARNS1' 'NRGN' 'NKX6-2' 'NELL2' 'KRT8' 'KCNC2' 'SYT1' 'CUX2' 'CABP1'
  'AACS' 'RTN1' 'HSPA2' 'PPP4R4' 'GLDN' 'HBA2' 'SLC5A11' 'AC009133.1'
  'PLLP' 'NDRG4' 'CALB2' 'KRT19' 'CNP' 'GFAP' 'NSF' 'ANKRD40' 'AATK'
  'MBP' 'CHGB' 'SNAP25' 'CST3' 'BCAS1' 'NDUFA7' 'PPP1R14A' 'CALM3'
  'RSPH14' 'NEFH' 'YWHAH' 'C21orf91' 'OLIG1' 'PCP4' 'MT-CO1' 'MT-CO2']
 ['CLSTN1' 'HPCA' 'LMO4' 'KCNA2' 'MEF2D' 'CADM3' 'RGS4' 'FAM84A'
  'SLC30A3' 'R3HDM1' 'ATP5MC3' 'EPHA4' 'ATP2B2' 'SYN2' 'CLDND1' 'TAGLN3'
  'TF' 'RAB6B' 'SERPINI1' 'PCDH7' 'HOPX' 'MAPK10' 'PPP3CA' 'UGT8' 'BASP1'
  'SELENOP' 'CARTPT' 'MAP1B' 'NREP' 'ANXA6' 'GABRA1' 'SNCB' 'TUBB2A'
  'NRN1' 'CAP2' 'TSPYL1' 'DGKB' 'AC018647.1' 'VSTM2A' 'TSPAN7' 'BEX1'
  'PLP1' 'FABP4' 'PDP1' 'GABBR2' 'SLC44A1' 'STXBP1' 'AL359091.1' 'ABCA2'
  'SLC1A2' 'FTH1' 'SCGB1D2' 'RTN3' 'C11orf87' 'CRYAB' 'THY1' 'HSPA8'
  'HEPACAM' 'SNCG' 'LIPA' 'LHPP' 'VAMP1' 'TUBA1B' 'TUBA1A' 'ATP5F1B'
  'FAM19A2' 'ATP2B1' 'HTR2A' 'PCDH8' 'HS6ST3' 'RNASE1' 'CALM1' 'ITPKA'
  'B2M' 'NPTN' 'CRYM' 'CACNG3' 'MT3' 'FA2H' 'CAMKK1' 'LRRC75A' 'KRT17'
  'SEPT4' 'AQP4' 'CNDP1' 'FKBP1A' 'LAMP5' 'VSTM2L' 'SLC12A5' 'CBLN4'
  'EEF1A2' 'AC005944.1' 'AC092069.1' 'HAPLN4' 'SCN1B' 'MAG' 'SLC17A7'
  'IGLC2' 'SULT4A1' 'MT-CO3']]

Iterative GeneCover via Combinatorial Optimization (SCIP Solver)

[11]:

# Obtain 200 marker genes via Iterative GeneCover with an incremental size of 100 and two iterations
genecover_markers_iterative = Iterative_GeneCover(incremental_sizes=[100,100], corr_mat=corr_mat, w=np.ones(corr_mat.shape[1]), solver = "SCIP")

Iteration 1
Best Gap:  0
Best Epsilon:  0.12958984375
Iteration  2
Best Gap:  0
Best Epsilon:  0.137158203125

[12]:

print("Iterative GeneCover markers: \n", orig_gene[genecover_markers_iterative])

Iterative GeneCover markers:
 [['PADI2' 'SYNC' 'WDR47' 'ATP1A1' 'ATP1B1' 'NME7' 'CNTN2' 'VSNL1' 'PREPL'
  'MDH1' 'PPP3R1' 'TMSB10' 'IGKC' 'MAL' 'ERMN' 'LRP2' 'GAD1' 'CHN1' 'CCK'
  'CLDN11' 'LDB2' 'SPP1' 'HHIP' 'TMEM144' 'GPM6A' 'SLC1A3' 'ENC1' 'EDIL3'
  'RHOBTB3' 'CXCL14' 'GABRB2' 'MOG' 'TPBG' 'ACTB' 'NDUFA4' 'RAPGEF5'
  'AQP1' 'YWHAG' 'SLC26A4-AS1' 'CD99' 'GJB1' 'NAP1L2' 'LAMP2' 'NEFM'
  'NEFL' 'STMN2' 'CALB1' 'ENPP2' 'CERCAM' 'OLFM1' 'PTGDS' 'SAA1' 'FOLH1'
  'MYRF' 'FTH1' 'SCGB2A2' 'PLA2G16' 'RTN3' 'CARNS1' 'HSPA8' 'NRGN'
  'NKX6-2' 'TUBA1B' 'KRT8' 'KCNC2' 'SYT1' 'CUX2' 'CABP1' 'AACS' 'RTN1'
  'HSPA2' 'PPP4R4' 'GLDN' 'NPTN' 'HBA1' 'CRYM' 'SLC5A11' 'AC009133.1'
  'PLLP' 'NDRG4' 'CALB2' 'KRT19' 'KRT17' 'GFAP' 'NSF' 'ANKRD40' 'AATK'
  'MBP' 'SNAP25' 'CST3' 'MAG' 'PPP1R14A' 'CALM3' 'RSPH14' 'NEFH' 'YWHAH'
  'OLIG1' 'PCP4' 'MT-CO1' 'MT-CO2']
 ['CLSTN1' 'HPCA' 'LMO4' 'KCNA2' 'CADM3' 'RGS4' 'HPCAL1' 'FAM84A'
  'SLC30A3' 'CALM2' 'R3HDM1' 'ATP5MC3' 'EPHA4' 'ATP2B2' 'SYN2' 'MOBP'
  'CLDND1' 'TAGLN3' 'TF' 'RAB6B' 'SERPINI1' 'FAM131A' 'PCDH7' 'UCHL1'
  'HOPX' 'PPP3CA' 'UGT8' 'BASP1' 'CARTPT' 'MAP1B' 'NREP' 'ANXA6' 'GABRA1'
  'SNCB' 'TUBB2A' 'NRN1' 'TSPYL1' 'PRKAR1B' 'DGKB' 'ANLN' 'VSTM2A'
  'PHKG1' 'TSPAN7' 'PLP1' 'CLU' 'FABP4' 'PDP1' 'DIRAS2' 'GABBR2'
  'SLC44A1' 'SLC31A2' 'STXBP1' 'AL359091.1' 'ABCA2' 'SCGB1D2' 'C11orf87'
  'CRYAB' 'THY1' 'HEPACAM' 'SNCG' 'LIPA' 'GOT1' 'SCD' 'LHPP' 'VAMP1'
  'FAM19A2' 'ATP2B1' 'DCLK1' 'HTR2A' 'PCDH8' 'HS6ST3' 'RNASE1' 'CALM1'
  'CHGA' 'ITPKA' 'B2M' 'CACNG3' 'MT3' 'FA2H' 'CAMKK1' 'LRRC75A' 'CNP'
  'TTYH2' 'AQP4' 'CNDP1' 'CHGB' 'SLC12A5' 'CBLN4' 'AC005944.1' 'ABHD8'
  'RAB3A' 'TMEM59L' 'HAPLN4' 'SCN1B' 'ATP1A3' 'SLC17A7' 'IGLC2' 'SULT4A1'
  'S100B' 'MT-CO3']]

Iterative GeneCover via Greedy Heuristics

[13]:

# Obtain 200 marker genes via Iterative GeneCover (Greedy Heuristics) with an incremental size of 100 and two iterations
genecover_markers_iterative = Iterative_GeneCover(incremental_sizes=[100,100], corr_mat=corr_mat, w=np.ones(corr_mat.shape[1]), solver= "Greedy")

Iteration 1
Best Gap:  0
Best Epsilon:  0.136181640625
Iteration  2
Best Gap:  0
Best Epsilon:  0.14765625

[14]:

print("Iterative GeneCover markers: \n", orig_gene[genecover_markers_iterative])

Iterative GeneCover markers:
 [['SNAP25' 'ERMN' 'ENC1' 'VSNL1' 'MAG' 'SCGB2A2' 'ATP1B1' 'RTN1' 'PTGDS'
  'OLFM1' 'MOG' 'MBP' 'FABP4' 'IGKC' 'CST3' 'CARNS1' 'GAD1' 'TMEM144'
  'KRT19' 'MT-CO1' 'NME7' 'UCHL1' 'PPP1R14A' 'CHN1' 'CLDN11' 'HBB'
  'HSPA2' 'CALB2' 'EDIL3' 'YWHAG' 'GJB1' 'NEFL' 'ENPP2' 'KRT8' 'RNASE1'
  'SLC5A11' 'GFAP' 'YWHAH' 'SYNC' 'RPL5' 'ATP1A1' 'HPCAL1' 'FAM84A'
  'PPP3R1' 'TMSB10' 'MOBP' 'LDB2' 'LIMCH1' 'SPP1' 'HHIP' 'SV2C' 'SLC12A2'
  'CXCL14' 'TUBB2A' 'ACTB' 'RAPGEF5' 'AQP1' 'PHKG1' 'SEMA3E' 'CHRDL1'
  'LAMP2' 'NEFM' 'CALB1' 'DIRAS2' 'SLC44A1' 'STXBP1' 'CERCAM' 'ABCA2'
  'FOLH1' 'FTH1' 'THY1' 'HSPA8' 'NRGN' 'NKX6-2' 'VAMP1' 'SYT1' 'CABP1'
  'AACS' 'HTR2A' 'SLAIN1' 'HS6ST3' 'GLDN' 'IQCK' 'GPRC5B' 'AC009133.1'
  'PLLP' 'NDRG4' 'RPL26' 'LRRC75A' 'KRT17' 'NSF' 'ANKRD40' 'SEPT4' 'AATK'
  'CHGB' 'LAMP5' 'NEFH' 'OLIG1' 'S100B' 'MT-CO3']
 ['TF' 'CCK' 'STMN2' 'NDUFA4' 'MDH1' 'PLP1' 'GPM6A' 'TUBA1B' 'HOPX'
  'SCGB1D2' 'MAL' 'NPTN' 'AC005944.1' 'MT-ND1' 'SAA1' 'CNP' 'RTN3'
  'IGLC2' 'SNCB' 'SLC1A2' 'CALM3' 'C11orf87' 'CRYAB' 'CNDP1' 'PCP4'
  'LMO4' 'CLDND1' 'GABRA1' 'TUBA1A' 'PLEKHH1' 'KLK6' 'ELOVL1' 'RGS4'
  'BASP1' 'RPL37' 'CARTPT' 'MAP1B' 'GABRB2' 'HLA-B' 'BEX1' 'LINC00844'
  'KCNC2' 'ATP2B1' 'MT3' 'HPCA' 'TMEM125' 'SV2A' 'CNTN2' 'SLC30A3'
  'CALM2' 'R3HDM1' 'ATP2B2' 'ARPP21' 'AC106707.1' 'SERPINI1' 'JCHAIN'
  'PARM1' 'G3BP2' 'PRDM8' 'PPP3CA' 'UGT8' 'CCDC152' 'SELENOP' 'HCN1'
  'HSP90AB1' 'PRKAR1B' 'ANLN' 'NACAD' 'GPR37' 'CLEC2L' 'TSPAN7' 'SYP'
  'NAP1L2' 'TCEAL6' 'PDP1' 'GABBR2' 'DNM1' 'MYRF' 'HEPACAM' 'FAM107B'
  'PIP4K2A' 'FRMPD2' 'HK1' 'OPALIN' 'GRIN2B' 'DCLK1' 'PCDH8' 'IGHA2'
  'IGHA1' 'ITPKA' 'ARPP19' 'SV2B' 'FA2H' 'CA10' 'BCAS1' 'EEF1A2' 'HAPLN4'
  'FTL' 'SLC17A7' 'SYNJ1']]

Marker Gene Selection Across Samples

Load the datasets (Sample #1515067, #151669, #151673)

[15]:

Adata = {}
data_dir = "..\\data\\DLPFC Full Samples"
file_names = os.listdir(data_dir)
across_samples_file_names = file_names[:3]
annotation_names = file_names[3:]
for j,  file in enumerate(across_samples_file_names):
    adata = sc.read_10x_h5(os.path.join(data_dir, file))
    adata.var_names_make_unique()
    layer_annotation = pd.read_csv(os.path.join(data_dir, annotation_names[j]), index_col=0)
    assert np.all(adata.obs.index.values == layer_annotation.index.values)
    adata.obs = layer_annotation
    sc.pp.filter_genes(adata, min_cells=100)
    sc.pp.normalize_total(adata, target_sum=1e4)
    sc.pp.log1p(adata)
    sc.pp.highly_variable_genes(adata, n_top_genes=10000)
    adata = adata[:, adata.var.highly_variable]
    Adata[file.split("_")[0]] = adata

for j, file in enumerate(across_samples_file_names):
    if j == 0:
        genes_across_samples = Adata[file.split("_")[0]].var_names.values
    else:
        genes_across_samples = np.intersect1d(genes_across_samples, Adata[file.split("_")[0]].var_names.values)

for key in across_samples_file_names:
    Adata[key.split("_")[0]] = Adata[key.split("_")[0]][:, genes_across_samples]

c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
c:\Users\An Wang\.conda\envs\vae-spatial\lib\site-packages\anndata\_core\anndata.py:1758: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")

Compute Correlation Matrices for Every Sample

[16]:

corr_mat_combine_across_samples = gene_gene_correlation([Adata[key.split("_")[0]].X.toarray() for key in across_samples_file_names])

GeneCover Marker Selection via Combinatorial Optimization Across Samples (Gurobi Solver)

[17]:

# Obtain 100 marker genes across samples
genecover_marker_across_samples = GeneCover(num_marker=100, corr_mat = corr_mat_combine_across_samples, w = np.ones(corr_mat_combine_across_samples.shape[1]), solver = "Gurobi")
print("GeneCover markers: \n", genes_across_samples[genecover_marker_across_samples])

Best Gap:  0
Best Epsilon:  0.10332946777343749
GeneCover markers:
 ['AC005944.1' 'AC009133.1' 'AC092069.1' 'ACTB' 'ALDOA' 'AP3B2' 'ATP1A1'
 'ATP1B1' 'ATP6V1B2' 'B2M' 'BASP1' 'CARTPT' 'CCK' 'CHN1' 'CLSTN2' 'CLSTN3'
 'CPLX1' 'CRYM' 'CST3' 'DIRAS2' 'DYNC1I1' 'ENC1' 'EPHA4' 'FABP4' 'G0S2'
 'GABRA1' 'GABRG2' 'GAD1' 'GAP43' 'GFAP' 'GPM6A' 'HBB' 'HOPX' 'IGHA1'
 'IGLC2' 'KLK6' 'KRT17' 'KRT19' 'KRT8' 'LMO4' 'MAGEE1' 'MAP1B' 'MBP'
 'MDH1' 'MICAL2' 'MOAP1' 'MOBP' 'MT-ND2' 'MT3' 'NDRG4' 'NEFH' 'NEFL'
 'NEFM' 'NELL2' 'NME7' 'NRGN' 'NSF' 'OAT' 'OLFM1' 'PAK1' 'PDP1' 'PHKG1'
 'PLIN1' 'PLP1' 'PPP3R1' 'RAB3A' 'REEP2' 'RPLP1' 'RTN1' 'RTN3' 'RTN4'
 'SCGB2A2' 'SCN1A' 'SH3BGRL2' 'SLC17A7' 'SLC1A2' 'SLC24A2' 'SLC39A10'
 'SNAP25' 'SNCB' 'SPP1' 'STMN1' 'STMN2' 'SYNC' 'SYNGR1' 'SYT1' 'SYT4'
 'TFF1' 'TMEM38A' 'TMEM59L' 'TMSB10' 'TUBA1A' 'TUBA1B' 'TUBB2A' 'UCHL1'
 'VPS35' 'VSNL1' 'WDR47' 'WDR7' 'YWHAG']

GeneCover Marker Selection via Combinatorial Optimization Across Samples (SCIP Solver)

[28]:

# Obtain 100 marker genes across samples
genecover_marker_across_samples = GeneCover(num_marker=100, corr_mat = corr_mat_combine_across_samples, w = np.ones(corr_mat_combine_across_samples.shape[1]), lambdaMax=.2,solver = "SCIP")
print("GeneCover markers: \n", genes_across_samples[genecover_marker_across_samples])

Best Gap:  0
Best Epsilon:  0.10302734375
GeneCover markers:
 ['AC005944.1' 'AC009133.1' 'AC092069.1' 'ACTB' 'ALDOA' 'AP3B2' 'APP'
 'ATL1' 'ATP1A1' 'ATP1B1' 'BASP1' 'CARNS1' 'CCK' 'CHN1' 'CLSTN2' 'CLSTN3'
 'CLU' 'CRYM' 'DYNC1I1' 'EIF1B' 'ENC1' 'EPHA4' 'FABP4' 'G0S2' 'GABRA1'
 'GAD1' 'GAP43' 'GFAP' 'GPM6A' 'GPRC5B' 'HBB' 'HOPX' 'HPCAL1' 'HSP90AB1'
 'IFI27' 'IGHA1' 'IGKC' 'KRT17' 'KRT19' 'KRT8' 'LDB2' 'LMO4' 'MAP1B' 'MBP'
 'MDH1' 'MICAL2' 'MOAP1' 'MOBP' 'MT-ND2' 'MT3' 'NDRG4' 'NECAB1' 'NEFH'
 'NEFL' 'NEFM' 'NELL2' 'NME7' 'NRGN' 'NRN1' 'NSF' 'OAT' 'OLFM1' 'PAK1'
 'PDP1' 'PFKP' 'PHKG1' 'PLIN1' 'PLP1' 'RAB3A' 'RPLP1' 'RTN1' 'RTN3'
 'SCGB2A2' 'SERPINF1' 'SH3BGRL2' 'SLC17A7' 'SLC1A2' 'SLC24A2' 'SNAP25'
 'SNCA' 'SNCB' 'SPP1' 'STMN1' 'STMN2' 'STXBP1' 'SYNC' 'SYNGR1' 'SYT1'
 'SYT4' 'TBR1' 'TFF1' 'TMEM38A' 'TMEM59L' 'TMSB10' 'TUBA1B' 'TUBB2A'
 'UCHL1' 'VSNL1' 'WDR47' 'YWHAG']

GeneCover Marker Selection via Greedy Heuristics Across Samples

[19]:

# Obtain 100 marker genes across samples via greedy heuristics
genecover_marker_across_samples_greedy = GeneCover(num_marker=100, corr_mat = corr_mat_combine_across_samples, w = np.ones(corr_mat_combine_across_samples.shape[1]), solver= "Greedy")
print("GeneCover markers: \n", genes_across_samples[genecover_marker_across_samples_greedy])

Best Gap:  0
Best Epsilon:  0.11860351562499999
GeneCover markers:
 ['SNAP25' 'PLP1' 'ENC1' 'SYT1' 'SCGB2A2' 'NRGN' 'VSNL1' 'CST3' 'ATP1B1'
 'MOG' 'TUBA1B' 'RTN1' 'SAA1' 'TMSB10' 'NEFL' 'GFAP' 'ERMN' 'IGKC' 'UCHL1'
 'KRT19' 'ACTB' 'MBP' 'CLDN11' 'MAP1B' 'OLFM1' 'PTGDS' 'MT-CO2' 'CARNS1'
 'CHN1' 'CLSTN2' 'RTN3' 'STMN1' 'TUBB2A' 'HBA1' 'NDUFA4' 'PPP1R14A'
 'SLC1A2' 'DIRAS2' 'AGT' 'ALDOA' 'ANXA6' 'ATP6V1C1' 'ATP6V1D' 'BASP1'
 'CDC37' 'CLDND1' 'CNDP1' 'CNTN2' 'COL1A2' 'COX6C' 'CREG2' 'CRYAB' 'DCLK1'
 'DYNC1I1' 'EFHD2' 'EPDR1' 'ETS2' 'FAM3C' 'FKBP1A' 'FTL' 'GLS' 'GPM6A'
 'HPRT1' 'HSPA2' 'HSPA8' 'HSPH1' 'KCNC2' 'KIFAP3' 'LAMP2' 'LDHA'
 'LINC00844' 'LMO4' 'MAG' 'MAGED1' 'MAL' 'MAP3K12' 'MDH1' 'MGP' 'MMACHC'
 'MT-ND2' 'NAP1L2' 'NAP1L5' 'NSF' 'OPALIN' 'PLIN1' 'PNMA8A' 'PRKAR1B'
 'PRKCE' 'RAB3A' 'RAB6A' 'RGS4' 'RPS27' 'SCG5' 'SCGB1D2' 'SERPINI1'
 'SLC1A3' 'SNCA' 'SNCG' 'SYN2' 'TFF1']

Iterative GeneCover via Combinatorial Optimization Across Samples (Gurobi Solver)

[22]:

# Obtain 200 marker genes via Iterative GeneCover with an incremental size of 100 and two iterations
genecover_marker_across_samples = Iterative_GeneCover(incremental_sizes=[100,100], corr_mat=corr_mat_combine_across_samples, w=np.ones(corr_mat_combine_across_samples.shape[1]), solver = "Gurobi")

Iteration 1
Best Gap:  0
Best Epsilon:  0.10332946777343749
Iteration  2
Best Gap:  0
Best Epsilon:  0.10310058593749999

[23]:

print("Iterative GeneCover markers: \n", genes_across_samples[genecover_marker_across_samples])

Iterative GeneCover markers:
 [['AC005944.1' 'AC009133.1' 'AC092069.1' 'ACTB' 'ALDOA' 'AP3B2' 'ATP1A1'
  'ATP1B1' 'ATP6V1B2' 'B2M' 'BASP1' 'CARTPT' 'CCK' 'CHN1' 'CLSTN2'
  'CLSTN3' 'CPLX1' 'CRYM' 'CST3' 'DIRAS2' 'DYNC1I1' 'ENC1' 'EPHA4'
  'FABP4' 'G0S2' 'GABRA1' 'GABRG2' 'GAD1' 'GAP43' 'GFAP' 'GPM6A' 'HBB'
  'HOPX' 'IGHA1' 'IGLC2' 'KLK6' 'KRT17' 'KRT19' 'KRT8' 'LMO4' 'MAGEE1'
  'MAP1B' 'MBP' 'MDH1' 'MICAL2' 'MOAP1' 'MOBP' 'MT-ND2' 'MT3' 'NDRG4'
  'NEFH' 'NEFL' 'NEFM' 'NELL2' 'NME7' 'NRGN' 'NSF' 'OAT' 'OLFM1' 'PAK1'
  'PDP1' 'PHKG1' 'PLIN1' 'PLP1' 'PPP3R1' 'RAB3A' 'REEP2' 'RPLP1' 'RTN1'
  'RTN3' 'RTN4' 'SCGB2A2' 'SCN1A' 'SH3BGRL2' 'SLC17A7' 'SLC1A2' 'SLC24A2'
  'SLC39A10' 'SNAP25' 'SNCB' 'SPP1' 'STMN1' 'STMN2' 'SYNC' 'SYNGR1'
  'SYT1' 'SYT4' 'TFF1' 'TMEM38A' 'TMEM59L' 'TMSB10' 'TUBA1A' 'TUBA1B'
  'TUBB2A' 'UCHL1' 'VPS35' 'VSNL1' 'WDR47' 'WDR7' 'YWHAG']
 ['ADD2' 'AK5' 'ANXA6' 'APBA2' 'APP' 'ATP6AP2' 'ATRNL1' 'BEX1' 'CABP1'
  'CACNB1' 'CALM3' 'CAMK2D' 'CAMK2N1' 'CHGA' 'CHGB' 'CLDN11' 'CLU' 'CMAS'
  'CNP' 'CPB1' 'CREG2' 'CRYAB' 'DCLK1' 'DNM1' 'DNM3' 'EEF1A1' 'EFHD2'
  'EIF4A2' 'EPDR1' 'ERMN' 'ETS2' 'FBXW7' 'G3BP2' 'GAPDH' 'GLS' 'GNB5'
  'HLA-B' 'HSP90AA1' 'HSP90AB1' 'HSPA8' 'HSPH1' 'IGKC' 'KCNK1' 'KIFAP3'
  'LRRC75A' 'MAPRE3' 'MFSD4A' 'MGP' 'MT-ND1' 'MUC1' 'NAP1L2' 'NDFIP1'
  'NELL1' 'NMNAT2' 'NPTN' 'NREP' 'NUAK1' 'NUDT4' 'OXR1' 'PCP4' 'PFKP'
  'PGK1' 'PIK3R1' 'PRKAR1A' 'PRKCE' 'PRPF19' 'PTGDS' 'PTPN5' 'RAB2A'
  'RAB3C' 'REEP1' 'RGS4' 'RPS12' 'RTN4RL2' 'SAA1' 'SCG5' 'SCGB1D2'
  'SCN2A' 'SERPINI1' 'SH3GL2' 'SLC1A3' 'SLC6A17' 'SNCG' 'SNX10' 'SPOCK2'
  'SRPK2' 'STXBP1' 'SV2A' 'SV2B' 'SYP' 'TAGLN3' 'TBR1' 'TF' 'THY1'
  'TOLLIP' 'TUBA4A' 'WASF1' 'WASL' 'YWHAB' 'YWHAH']]

Iterative GeneCover via Combinatorial Optimization Across Samples (SCIP Solver)

[30]:

# Obtain 200 marker genes via Iterative GeneCover with an incremental size of 100 and two iterations
genecover_marker_across_samples = Iterative_GeneCover(incremental_sizes=[100,100], corr_mat=corr_mat_combine_across_samples, w=np.ones(corr_mat_combine_across_samples.shape[1]), lambdaMax=.2 ,solver = "SCIP")

Iteration 1
Best Gap:  0
Best Epsilon:  0.10302734375
Iteration  2
Best Gap:  0
Best Epsilon:  0.102752685546875

[32]:

print("Iterative GeneCover markers: \n", genes_across_samples[genecover_marker_across_samples])

Iterative GeneCover markers:
 [['AC005944.1' 'AC009133.1' 'AC092069.1' 'ACTB' 'ALDOA' 'AP3B2' 'APP'
  'ATL1' 'ATP1A1' 'ATP1B1' 'BASP1' 'CARNS1' 'CCK' 'CHN1' 'CLSTN2'
  'CLSTN3' 'CLU' 'CRYM' 'DYNC1I1' 'EIF1B' 'ENC1' 'EPHA4' 'FABP4' 'G0S2'
  'GABRA1' 'GAD1' 'GAP43' 'GFAP' 'GPM6A' 'GPRC5B' 'HBB' 'HOPX' 'HPCAL1'
  'HSP90AB1' 'IFI27' 'IGHA1' 'IGKC' 'KRT17' 'KRT19' 'KRT8' 'LDB2' 'LMO4'
  'MAP1B' 'MBP' 'MDH1' 'MICAL2' 'MOAP1' 'MOBP' 'MT-ND2' 'MT3' 'NDRG4'
  'NECAB1' 'NEFH' 'NEFL' 'NEFM' 'NELL2' 'NME7' 'NRGN' 'NRN1' 'NSF' 'OAT'
  'OLFM1' 'PAK1' 'PDP1' 'PFKP' 'PHKG1' 'PLIN1' 'PLP1' 'RAB3A' 'RPLP1'
  'RTN1' 'RTN3' 'SCGB2A2' 'SERPINF1' 'SH3BGRL2' 'SLC17A7' 'SLC1A2'
  'SLC24A2' 'SNAP25' 'SNCA' 'SNCB' 'SPP1' 'STMN1' 'STMN2' 'STXBP1' 'SYNC'
  'SYNGR1' 'SYT1' 'SYT4' 'TBR1' 'TFF1' 'TMEM38A' 'TMEM59L' 'TMSB10'
  'TUBA1B' 'TUBB2A' 'UCHL1' 'VSNL1' 'WDR47' 'YWHAG']
 ['AC011603.2' 'ADD2' 'AK5' 'APBA2' 'ATP1B2' 'ATP5F1A' 'ATP6AP2'
  'ATP6V1B2' 'BEX1' 'CABP1' 'CACNB1' 'CALM3' 'CAMK2D' 'CAP2' 'CARTPT'
  'CDKN2D' 'CHGA' 'CHGB' 'CLDN11' 'CNP' 'CREG2' 'CRYAB' 'CST3' 'CX3CL1'
  'DCLK1' 'DNAJB6' 'DNAJC6' 'DNM1' 'DNM3' 'EFHD2' 'EHD3' 'EPDR1' 'ERMN'
  'FBXW7' 'FKBP1A' 'G3BP2' 'GABRG2' 'GAPDH' 'GGCX' 'GLS' 'GOT1' 'HLA-B'
  'HMGCS1' 'HSP90AA1' 'HSPA8' 'IGLC2' 'KCNK1' 'KIFAP3' 'LRRC75A' 'MAPRE3'
  'MAPT' 'MGP' 'MTRNR2L8' 'MUC1' 'NAP1L2' 'NCALD' 'NCOA7' 'NDFIP1'
  'NMNAT2' 'NPTN' 'NUAK1' 'NUDT4' 'OXR1' 'PCP4' 'PGK1' 'PI4KA' 'PIK3R1'
  'PNMA8A' 'PPP3R1' 'PREPL' 'PRKCE' 'PRNP' 'PRPF19' 'PTGDS' 'PTPN5'
  'RAB15' 'RAB3C' 'REEP1' 'RPS12' 'RTN4' 'SAA1' 'SCGB1D2' 'SCN2A' 'SCN3B'
  'SERPINI1' 'SH3GL2' 'SNX10' 'SV2A' 'SV2B' 'SYP' 'TAGLN3' 'TF' 'THY1'
  'TOLLIP' 'TRIM37' 'TUBA4A' 'VPS35' 'WASF1' 'YWHAB' 'YWHAH']]

Iterative GeneCover via Greedy Heuristics Across Samples

[ ]:

# Obtain 200 marker genes via Iterative GeneCover with an incremental size of 100 and two iterations
genecover_marker_across_samples_greedy = genecover_markers_iterative = Iterative_GeneCover(incremental_sizes=[100,100], corr_mat=corr_mat_combine_across_samples, w=np.ones(corr_mat_combine_across_samples.shape[1]),lambdaMin = .08, lambdaMax = .2, solver= "Greedy")

Iteration 1
Best Gap:  0
Best Epsilon:  0.1185546875
Iteration  2
Best Gap:  0
Best Epsilon:  0.11213867187500001

[ ]:

print("Iterative GeneCover markers: \n", genes_across_samples[genecover_marker_across_samples_greedy])

Iterative GeneCover markers:
 [['SNAP25' 'PLP1' 'ENC1' 'SYT1' 'SCGB2A2' 'NRGN' 'VSNL1' 'CST3' 'ATP1B1'
  'MOG' 'TUBA1B' 'RTN1' 'SAA1' 'TMSB10' 'NEFL' 'GFAP' 'ERMN' 'IGKC'
  'UCHL1' 'KRT19' 'ACTB' 'MBP' 'CLDN11' 'MAP1B' 'OLFM1' 'PTGDS' 'MT-CO2'
  'CARNS1' 'CHN1' 'CLSTN2' 'RTN3' 'STMN1' 'TUBB2A' 'HBA1' 'NDUFA4'
  'PPP1R14A' 'SLC1A2' 'DIRAS2' 'AGT' 'ALDOA' 'ANXA6' 'ATP6V1C1' 'ATP6V1D'
  'BASP1' 'CDC37' 'CLDND1' 'CNDP1' 'CNTN2' 'COL1A2' 'COX6C' 'CREG2'
  'DCLK1' 'DYNC1I1' 'EFHD2' 'EPDR1' 'ETS2' 'FAM3C' 'FKBP1A' 'FTL' 'GLS'
  'GPM6A' 'HPRT1' 'HSPA2' 'HSPA8' 'HSPH1' 'KCNC2' 'KIFAP3' 'KRT18'
  'LAMP2' 'LDHA' 'LINC00844' 'LMO4' 'MAG' 'MAGED1' 'MAL' 'MAP3K12' 'MDH1'
  'MGP' 'MMACHC' 'MT-ND2' 'NAP1L2' 'NAP1L5' 'NSF' 'OPALIN' 'PLIN1'
  'PNMA8A' 'PRKAR1B' 'PRKCE' 'RAB3A' 'RAB6A' 'RGS4' 'RPS27' 'SCG5'
  'SCGB1D2' 'SERPINI1' 'SLC1A3' 'SNCA' 'SNCG' 'SYN2' 'TFF1']
 ['YWHAH' 'TF' 'STMN2' 'YWHAG' 'SLC24A2' 'CCK' 'MT3' 'MUC1' 'NEFM' 'NME7'
  'FABP4' 'MOBP' 'HOPX' 'ATP6V1B2' 'IGLC2' 'PCP4' 'SLC17A7' 'PFKP' 'TBR1'
  'CAP2' 'APP' 'GABRA1' 'NKX6-2' 'SYP' 'TAGLN3' 'SH3GL2' 'GAP43' 'CRYM'
  'AC005944.1' 'CNP' 'THY1' 'MOAP1' 'NELL1' 'SYT4' 'NAPB' 'SLC30A3'
  'OXR1' 'SPP1' 'STXBP1' 'GABRG2' 'MICAL2' 'NDRG4' 'CHL1' 'ENPP2' 'FABP3'
  'IGHA1' 'ITFG1' 'KRT8' 'MT-CO1' 'PAK1' 'SCN2A' 'SSTR2' 'TUBA4A' 'CHGB'
  'CLU' 'FAM49A' 'HSP90AA1' 'LRRC75A' 'NELL2' 'NUAK1' 'PGM2L1'
  'AC011603.3' 'ACOT13' 'ACOT7' 'APBA2' 'APOE' 'ATL1' 'ATP1B2' 'ATP6V1A'
  'BNIP3' 'CPLX1' 'CRYAB' 'CTNNB1' 'EIF4A2' 'ENO2' 'GHITM' 'GLUL'
  'IPCEF1' 'KIF21A' 'KLC1' 'MAP2K1' 'MLLT11' 'MT-ND1' 'MT-ND3' 'MYRF'
  'PCLO' 'PCSK2' 'PEG3' 'PIK3R1' 'S100A11' 'SEPT4' 'SLC6A17' 'STX1A'
  'SYNGR1' 'SYNGR3' 'TMEM144' 'TMEM14A' 'TOLLIP' 'TUBB' 'VAMP1']]