def RunClusterAcc(dataset1, dataset2, gene_dataset, plotname):
cluster1 = KMeans(len(dataset1.cell_types))
cluster2 = KMeans(len(dataset2.cell_types))
latent1 = np.genfromtxt('../harmonization/Seurat_data/' + plotname +
'.1.CCA.txt')
latent2 = np.genfromtxt('../harmonization/Seurat_data/' + plotname +
'.2.CCA.txt')
latent, batch_indices, labels, keys, stats = run_model('readSeurat',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_seurat = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
latent, batch_indices, labels, keys, stats = run_model('MNN',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_MNN = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
latent, batch_indices, labels, keys, stats = run_model('PCA',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_PCA = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
dataset1, dataset2, gene_dataset = SubsetGenes(dataset1, dataset2,
gene_dataset, plotname)
latent1, _, _, _, _ = run_model('vae',
dataset1,
0,
0,
filename=plotname,
rep='vae1')
latent2, _, _, _, _ = run_model('vae',
dataset2,
0,
0,
filename=plotname,
rep='vae2')
latent, batch_indices, labels, keys, stats = run_model('vae',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_scvi = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
latent, batch_indices, labels, keys, stats = run_model('vae_nb',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_scvi_nb = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
latent, batch_indices, labels, keys, stats = run_model('scanvi1',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_scanvi1 = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
latent, batch_indices, labels, keys, stats = run_model('scanvi2',
gene_dataset,
dataset1,
dataset2,
filename=plotname)
res_scanvi2 = computeARI(latent1, latent2, latent, cluster1, cluster2,
batch_indices)
res = [
res_scvi, res_scvi_nb, res_scanvi1, res_scanvi2, res_seurat, res_MNN,
res_PCA
]
res = np.asarray(res)
np.savetxt("%s.clusterScore.csv" % (plotname), res, "%.4f", ',')
from scvi.dataset.dataset import GeneExpressionDataset
from scvi.harmonization.utils_chenling import SubsetGenes
import pickle as pkl
f = open('../%s/gene_dataset.pkl'%plotname, 'rb')
all_dataset, dataset1, dataset2 = pkl.load(f)
f.close()
all_dataset = GeneExpressionDataset.concat_datasets(dataset1,dataset2)
dataset1, dataset2, gene_dataset = SubsetGenes(dataset1, dataset2, all_dataset, plotname)
import time
from scvi.harmonization.utils_chenling import run_model
start = time.time()
latent, batch_indices, labels, keys, stats = run_model('scmap', gene_dataset, dataset1, dataset2,filename=plotname)
end = time.time()
print( end - start)
batch = gene_dataset.batch_indices.ravel()
labels = gene_dataset.labels.ravel()
scaling_factor = gene_dataset.X.mean(axis=1)
norm_X = gene_dataset.X / scaling_factor.reshape(len(scaling_factor), 1)
index_0 = np.where(batch == 0)[0]
index_1 = np.where(batch == 1)[0]
X1 = np.log(1 + norm_X[index_0])
X2 = np.log(1 + norm_X[index_1])
coral = CORAL()
pbmc2.filter_cell_types(newCellType)
gene_dataset = GeneExpressionDataset.concat_datasets(pbmc, pbmc2)
# _,_,_,_,_ = run_model('writedata', gene_dataset, pbmc, pbmc2,filename=plotname+'.'
# +celltype1.replace(' ','')+'.'
# +celltype2.replace(' ',''))
rmCellTypes = '.' + celltype1.replace(
' ', '') + '.' + celltype2.replace(' ', '')
latent1 = np.genfromtxt('../harmonization/Seurat_data/' +
plotname + rmCellTypes.replace(' ', '') +
'.1.CCA.txt')
latent2 = np.genfromtxt('../harmonization/Seurat_data/' +
plotname + rmCellTypes.replace(' ', '') +
'.2.CCA.txt')
latent, batch_indices, labels, keys, stats = run_model(
'readSeurat',
gene_dataset,
pbmc,
pbmc2,
filename=plotname + rmCellTypes.replace(' ', ''))
acc, cell_type = KNNpurity(latent1, latent2, latent,
batch_indices.ravel(), labels, keys)
f.write('Seurat' + '\t' + rmCellTypes +
("\t%.4f" * 8 + "\t%s" * 8 + "\n") %
tuple(list(acc) + list(cell_type)))
be, temp1 = BEbyType(keys, latent, labels, batch_indices,
celltype1)
g.write('Seurat' + '\t' + rmCellTypes +
("\t%.4f" * 8 + "\t%s" * 8 + "\n") %
tuple(be + list(temp1)))
plotUMAP(latent, plotname, 'Seurat', gene_dataset.cell_types,
rmCellTypes, gene_dataset.batch_indices.ravel())
pbmc, pbmc2, gene_dataset = SubsetGenes(