Machine learning-based investigation of the cancer protein secretory pathway

Abstract

Deregulation of the protein secretory pathway (PSP) is linked to many hallmarks of cancer, such as promoting tissue invasion and modulating cell-cell signaling. The collection of secreted proteins processed by the PSP, known as the secretome, is often studied due to its potential as a reservoir of tumor biomarkers. However, there has been less focus on the protein components of the secretory machinery itself. We therefore investigated the expression changes in secretory pathway components across many different cancer types. Specifically, we implemented a dual approach involving differential expression analysis and machine learning to identify PSP genes whose expression was associated with key tumor characteristics: mutation of p53, cancer status, and tumor stage. Eight different machine learning algorithms were included in the analysis to enable comparison between methods and to focus on signals that were robust to algorithm type. The machine learning approach was validated by identifying PSP genes known to be regulated by p53, and even outperformed the differential expression analysis approach. Among the different analysis methods and cancer types, the kinesin family members KIF20A and KIF23 were consistently among the top genes associated with malignant transformation or tumor stage. However, unlike most cancer types which exhibited elevated KIF20A expression that remained relatively constant across tumor stages, renal carcinomas displayed a more gradual increase that continued with increasing disease severity. Collectively, our study demonstrates the complementary nature of a combined differential expression and machine learning approach for analyzing gene expression data, and highlights key PSP components relevant to features of tumor pathophysiology that may constitute potential therapeutic targets.

Integrative Pan-Cancer Analysis of KIF15 Reveals Its Diagnosis and Prognosis Value in Nasopharyngeal Carcinoma

“…The well-known cancer suppressor gene, p53, closely controls various cellular signals involved in the cell-cycle, apoptosis, and senescence ( 39 ). A study based on machine learning showed that KIF20A and KIF23 were regulated by p53 and correlated with malignant transformation and tumor stage ( 40 ). Further, KIF15 knockdown strongly enhanced the expression of p53 and p21 protein in breast cancer cells ( 37 ).…”

Prostate cancer in omics era

“…Recently, Saghaleyni and colleagues developed a module with eight machine learning algorithms for analyzing secreted protein profiles using gene expression data from thousands of normal human tissue and tumor samples. KIF20A and KIF23 , members of the kinesin family, were consistently among the top genes linked to malignant transformation [ 96 ]. Multi-Omic Graph Convolutional NETworks (MOGONET) effectively categorize multi-omic data, by both omics-specific learning and cross-omics correlation learning.…”

Identification of secretory pathway‐related genes based on Random Forest algorithm to predict the prognosis and immunotherapy response of hepatocellular carcinoma

“…14 Several key secretory pathway components and kinase or kinase-like protein related to pathophysiological characteristics of tumors that may be potential therapeutic targets have been reported. 15,16 Future research on carcinogenesis will be outlined and anti-tumor therapy will be improved with the identification of novel targets connected to the secretory pathway. This work investigated the molecular alterations and features of secretory pathway expression, categorized HCC according to the expression of gene in secretory pathway and created a gene signature to assess the effectiveness of chemotherapy and immunotherapy and predict the prognosis of HCC.…”

“…Cancer secretome, which refers to an entire protein collection secreted via distinct secretory pathways by cancer cells, could represent potential therapeutic targets for a variety of cancers and tumor biomarkers 14 . Several key secretory pathway components and kinase or kinase‐like protein related to pathophysiological characteristics of tumors that may be potential therapeutic targets have been reported 15,16 . Future research on carcinogenesis will be outlined and anti‐tumor therapy will be improved with the identification of novel targets connected to the secretory pathway.…”