A systematic comparison of data- and knowledge-driven approaches to disease subtype discovery

Comprehensive Analysis of the Oncogenic Role of Targeting Protein for Xklp2 (TPX2) in Human Malignancies

Mitosis and spindle assembly require the microtubule-associated protein Xenopus kinesin-like protein 2 (TPX2). Although TPX2 is highly expressed in several malignant tumor forms, little is known about its role in cancer. In this study, we performed the gene set enrichment analysis of TPX2 in 33 types of cancers and an extensive pan-cancer bioinformatic analysis using prognosis, tumor mutational burdens, microsatellite instability, tumor microenvironment, and immune cell infiltration data. According to the differential expression study, TPX2 was found to be overexpressed across all studied cancer types. Based on the survival analysis, increased TPX2 expression was associated with a poor prognosis for most cancers. The TPX2 expression level was confirmed to correlate with the clinical stage, microsatellite instability, and tumor mutational burden across all cancer types. Furthermore, TPX2 expression has been linked to tumor microenvironments and immune cell infiltration, particularly in bladder urothelial carcinoma, liver hepatocellular carcinoma, lung adenocarcinoma, stomach adenocarcinoma, and uterine corpus endometrial carcinoma. Finally, the gene set enrichment analysis implicated TPX2 in the regulation of aminoacyl tRNA biosynthesis, which is the most important tumor cell cycle signaling pathway. This comprehensive pan-cancer analysis shows that TPX2 is a prognostic molecular biomarker for most cancers and suggests its potential as an effective therapeutic target for the treatment of these diseases.

Machine learning in endoscopic ultrasonography and the pancreas: The new frontier?

Pancreatic diseases have a substantial burden on society which is predicted to increase further over the next decades. Endoscopic ultrasonography (EUS) remains the best available diagnostic method to assess the pancreas, however, there remains room for improvement. Artificial intelligence (AI) approaches have been adopted to assess pancreatic diseases for over a decade, but this methodology has recently reached a new era with the innovative machine learning algorithms which can process, recognize, and label endosonographic images. Our review provides a targeted summary of AI in EUS for pancreatic diseases. Included studies cover a wide spectrum of pancreatic diseases from pancreatic cystic lesions to pancreatic masses and diagnosis of pancreatic cancer, chronic pancreatitis, and autoimmune pancreatitis. For these, AI models seemed highly successful, although the results should be evaluated carefully as the tasks, datasets and models were greatly heterogenous. In addition to use in diagnostics, AI was also tested as a procedural real-time assistant for EUS-guided biopsy as well as recognition of standard pancreatic stations and labeling anatomical landmarks during routine examination. Studies thus far have suggested that the adoption of AI in pancreatic EUS is highly promising and further opportunities should be explored in the field.