Simulated Cancer Omics dataset using InterSim CRAN package

Import packages and IntegrAO code

import numpy as np
import pandas as pd
import snf
from sklearn.cluster import spectral_clustering
from sklearn.metrics import v_measure_score
import matplotlib.pyplot as plt

import sys
import os
import argparse
import torch

import umap
from sklearn.model_selection import train_test_split

# Add the parent directory of "integrao" to the Python path
module_path = os.path.abspath(os.path.join('../'))
if module_path not in sys.path:
    sys.path.append(module_path)

from integrao.dataset import GraphDataset
from integrao.main import dist2
from integrao.integrater import integrao_integrater

Set hyperparameters

torch.cuda.is_available()

True

# Hyperparameters
neighbor_size = 20
embedding_dims = 64
fusing_iteration = 30
normalization_factor = 1.0
alighment_epochs = 1000
beta = 1.0
mu = 0.5


dataset_name = 'cancer omics'
cluster_number = 15

Read data

testdata_dir = os.path.join(module_path, "data/omics/")

methyl_ = os.path.join(testdata_dir, "omics1.txt")
expr_ = os.path.join(testdata_dir, "omics2.txt")
protein_ = os.path.join(testdata_dir, "omics3.txt")
truelabel = os.path.join(testdata_dir, "clusters.txt")


methyl = pd.read_csv(methyl_, index_col=0, delimiter="\t")
expr = pd.read_csv(expr_, index_col=0, delimiter="\t")
protein = pd.read_csv(protein_, index_col=0, delimiter="\t")
truelabel = pd.read_csv(truelabel, index_col=0, delimiter="\t")

methyl = np.transpose(methyl)
expr = np.transpose(expr)
protein = np.transpose(protein)
print(methyl.shape)
print(expr.shape)
print(protein.shape)
print(truelabel.shape)
print("finish loading data!")

(500, 367)
(500, 131)
(500, 160)
(500, 2)
finish loading data!

KNN clustering for clustering cnv, meth and mirna raw features as baseline

from sklearn.cluster import KMeans
from sklearn.metrics import v_measure_score

# cnv
kmeans = KMeans(n_clusters=cluster_number, random_state=42, n_init='auto').fit(methyl.values)
labels = kmeans.labels_
true_labels = truelabel['cluster.id'].tolist()
score = v_measure_score(true_labels, labels)
print("methyl raw features kmeans clustering NMI: {}".format(score))

# meth
kmeans = KMeans(n_clusters=cluster_number, random_state=42, n_init='auto').fit(expr.values)
labels = kmeans.labels_
true_labels = truelabel['cluster.id'].tolist()
score = v_measure_score(true_labels, labels)
print("expr raw features kmeans clustering NMI: {}".format(score))

# mirna
kmeans = KMeans(n_clusters=cluster_number, random_state=42, n_init='auto').fit(protein.values)
labels = kmeans.labels_
true_labels = truelabel['cluster.id'].tolist()
score = v_measure_score(true_labels, labels)
print("protein raw features kmeans clustering NMI: {}".format(score))

methyl raw features kmeans clustering NMI: 0.4092647077881033
expr raw features kmeans clustering NMI: 0.7651290634507789
protein raw features kmeans clustering NMI: 0.868987364136353

Random sub-sample the omics dataset to create an incomplete dataset

ratio = 0.7

full_indices = range(len(methyl))
unique_indices, common_indices = train_test_split(full_indices, test_size=ratio)

w1w2_indices, w3_indices = train_test_split(unique_indices, test_size=0.33)
w1_indices, w2_indices = train_test_split(w1w2_indices, test_size=0.5)

w1_full_indices = common_indices + w1_indices
w2_full_indices = common_indices + w2_indices
w3_full_indices = common_indices + w3_indices

methyl_temp = methyl.iloc[w1_full_indices]
expr_temp = expr.iloc[w2_full_indices]
protein_temp = protein.iloc[w3_full_indices]

IntegrAO integration

# Initialize integrater
integrater = integrao_integrater(
    [methyl_temp, expr_temp, protein_temp],
    dataset_name,
    neighbor_size=neighbor_size,
    embedding_dims=embedding_dims,
    fusing_iteration=fusing_iteration,
    normalization_factor=normalization_factor,
    alighment_epochs=alighment_epochs,
    beta=beta,
    mu=mu,
)
# data indexing
fused_networks = integrater.network_diffusion()
embeds_final, S_final, model = integrater.unsupervised_alignment()

labels = spectral_clustering(S_final, n_clusters=cluster_number)

true_labels = truelabel.sort_values('subjects')['cluster.id'].tolist()

score_all = v_measure_score(true_labels, labels)
print("IntegrAO for clustering union 500 samples NMI score: ", score_all)

Start indexing input expression matrices!
Common sample between view0 and view1: 350
Common sample between view0 and view2: 350
Common sample between view1 and view2: 350
Neighbor size: 20
Start applying diffusion!
Diffusion ends! Times: 15.066577911376953s
Starting unsupervised exmbedding extraction!
Dataset 0: (400, 367)
Dataset 1: (400, 131)
Dataset 2: (400, 160)
epoch 0: loss 30.054597854614258, align_loss:0.775736
epoch 100: loss 20.709983825683594, align_loss:0.151950
epoch 200: loss 1.092381477355957, align_loss:0.082980
epoch 300: loss 1.0914002656936646, align_loss:0.082236
epoch 400: loss 1.0902798175811768, align_loss:0.081427
epoch 500: loss 1.0890438556671143, align_loss:0.080614
epoch 600: loss 1.0877360105514526, align_loss:0.079772
epoch 700: loss 1.0863633155822754, align_loss:0.078964
epoch 800: loss 1.0849252939224243, align_loss:0.078173
epoch 900: loss 1.0834227800369263, align_loss:0.077401
Manifold alignment ends! Times: 56.45441031455994s
IntegrAO for clustering union 500 samples NMI score:  0.9862486593875475

UMAP plot of the integrated embeddings

import umap

# umap visualization of embeddings with hue as labels
reducer = umap.UMAP(n_neighbors=30, min_dist=1.0, spread=1.0)
embedding = reducer.fit_transform(embeds_final.values)
plt.figure(figsize=(10, 10))
plt.scatter(
    embedding[:, 0],
    embedding[:, 1],
    c=true_labels,
    s=6,
    cmap="Spectral",
    alpha=1.0,
)
plt.setp(plt.gca(), xticks=[], yticks=[])
plt.title("UMAP projection of the dataset", fontsize=24)
plt.colorbar()
plt.show()