Unsupervised Clustering of Immunohistochemical Markers to Define High-Risk Endometrial Cancer

Abstract

Considerable heterogeneity exists in outcomes of early endometrial cancer (EC) according to the type but also the histological grading. Our goal was to describe the immunohistochemical profiles of type I EC according to grades and type II EC, to identify groups of interacting proteins using principal component analysis (PCA) and unsupervised clustering. We studied 13 immunohistochemical markers (steroid receptors, pro/anti-apoptotic proteins, metalloproteinases (MMP) and tissue inhibitor of metalloproteinase (TIMP), and CD44 isoforms known for their role in endometrial pathology. Co-expressed proteins associated with the type, grade and outcome of EC were determined by PCA and unsupervised clustering. PCA identified three functional groups of proteins from 43 tissue samples (38 type I and 5 type II EC): the first was characterized by p53 expression; the second by MMPs, bcl-2, PR B and CD44v6; and the third by ER alpha, PR A, TIMP-2 and CD44v3. Unsupervised clustering found two main clusters of proteins, with both type I grade 3 and type II EC exhibiting the same cluster profile. PCA and unsupervised clustering of immunohistochemical markers in EC contribute to a better comprehension and classification of the disease.

Collagen IV and laminin‐1 expression in embryonic mouse lens using principal components analysis technique

“…The PCA method is used to determine which proteins are most representative of all the variability of the data and the usual process is to mark the tissues only with them and not with the others. This reduces the number of samples to be used and the types of labelling (Gry et al ., ; Sobhani et al ., ; Laas et al ., ). In other cases, the immunohistochemical images are used to illustrate the morphological patterns characteristic of the expression of the molecular markers (Moro et al ., ). Enhancement of the immunohistochemistry imaging contrast by performing the PCA analysis on the RGB image taken with a colour CCD camera, together with automatic classification algorithms or Machine Learning (Loukas et al ., ; Khorshed, ; Sarnecki et al ., ; Komura & Ishikawa, ; Van Eycke et al ., ). The degree of marking of an immunohistochemical image is usually done through a subjective numerical ‘scoring’ that assigns a ‘score’ to each image.…”

“…The PCA method is used to determine which proteins are most representative of all the variability of the data and the usual process is to mark the tissues only with them and not with the others. This reduces the number of samples to be used and the types of labelling (Gry et al ., ; Sobhani et al ., ; Laas et al ., ). In other cases, the immunohistochemical images are used to illustrate the morphological patterns characteristic of the expression of the molecular markers (Moro et al ., ).…”

In‐depth comparative analysis of Illumina® MiSeq run metrics: Development of a wet‐lab quality assessment tool

“…The main drawbacks of using K ‐means clustering alone is associated with two well‐established problems such as (a) defining a priori the number of clusters to use and (b) visualizing the obtained clusters in several dimensions. Thus, a typical solution is to preprocess the data using PCA by mapping the data into a new feature space (Laas, Ballester, Cortez, Graesslin, & Daraї, ). Afterwards, the k ‐means algorithm is applied to the data in the feature space.…”

Diagnosis Delay and Assessment of the Quality of Life of Patients with Endometriosis Using the Endometriosis Health Profile 5 Questionnaire in a Sub-Saharan Population

“…This incidence reaches 12% to 45% in infertile women [6] [7]. However, it should be emphasized that these epidemiological data emanate from developed countries benefiting easily from advanced imaging modalities (transvaginal ultrasound and MRI) [8]- [12]. Conversely, in sub-Saharan countries, diagnostic tools for endometriosis are more limited thus explaining the insufficient epidemiological data, especially on diagnostic delay, impact on quality of life and fertility management [4] [9]- [13].…”