Comparison of Vision Transformers and Convolutional Neural Networks in Medical Image Analysis: A Systematic Review.

In the rapidly evolving field of medical image analysis utilizing artificial intelligence (AI), the selection of appropriate computational models is critical for accurate diagnosis and patient care. This literature review provides a comprehensive comparison of vision transformers (ViTs) and convolutional neural networks (CNNs), the two leading techniques in the field of deep learning in medical imaging. We conducted a survey systematically. Particular attention was given to the robustness, computational efficiency, scalability, and accuracy of these models in handling complex medical datasets. The review incorporates findings from 36 studies and indicates a collective trend that transformer-based models, particularly ViTs, exhibit significant potential in diverse medical imaging tasks, showcasing superior performance when contrasted with conventional CNN models. Additionally, it is evident that pre-training is important for transformer applications. We expect this work to help researchers and practitioners select the most appropriate model for specific medical image analysis tasks, accounting for the current state of the art and future trends in the field.

Multi-omics and clustering analyses reveal the mechanisms underlying unmet needs for patients with lung adenocarcinoma and identify potential therapeutic targets.

BACKGROUND: The cancer genome contains several driver mutations. However, in some cases, no known drivers have been identified; these remaining areas of unmet needs, leading to limited progress in cancer therapy. Whole-genome sequencing (WGS) can identify non-coding alterations associated with the disease. Consequently, exploration of non-coding regions using WGS and other omics data such as ChIP-sequencing (ChIP-seq) to discern novel alterations and mechanisms related to tumorigenesis have been attractive these days. METHODS: Integrated multi-omics analyses, including WGS, ChIP-seq, DNA methylation, and RNA-sequencing (RNA-seq), were conducted on samples from patients with non-clinically actionable genetic alterations (non-CAGAs) in lung adenocarcinoma (LUAD). Second-level cluster analysis was performed to reinforce the correlations associated with patient survival, as identified by RNA-seq. Subsequent differential gene expression analysis was performed to identify potential druggable targets. RESULTS: Differences in H3K27ac marks in non-CAGAs LUAD were found and confirmed by analyzing RNA-seq data, in which mastermind-like transcriptional coactivator 2 (MAML2) was suppressed. The down-regulated genes whose expression was correlated to MAML2 expression were associated with patient prognosis. WGS analysis revealed somatic mutations associated with the H3K27ac marks in the MAML2 region and high levels of DNA methylation in MAML2 were observed in tumor samples. The second-level cluster analysis enabled patient stratification and subsequent analyses identified potential therapeutic target genes and treatment options. CONCLUSIONS: We overcome the persistent challenges of identifying alterations or driver mutations in coding regions related to tumorigenesis through a novel approach combining multi-omics data with clinical information to reveal the molecular mechanisms underlying non-CAGAs LUAD, stratify patients to improve patient prognosis, and identify potential therapeutic targets. This approach may be applicable to studies of other cancers with unmet needs.

Method to Purify Phase-Separated p62 Bodies Using Fluorescence-Activated Particle Sorting.

p62 bodies are ubiquitin-positive cytoplasmic condensates formed by liquid-liquid phase separation. They are targeted by selective autophagy and play important roles in intracellular quality control and stress responses. However, little is known about their constituents. In this chapter, we describe a method for purifying p62 bodies using fluorescence-activated particle sorting. This method contributes to the identification of novel components of p62 bodies under various physiological and stress conditions.

Excellent concordance of the molecular classification between preoperative biopsy and final hysterectomy in endometrial carcinoma.

OBJECTIVE: The 2023 International Federation of Gynecology and Obstetrics classification with molecular classification shows superior discriminatory ability compared to staging systems lacking molecular data. However, the accuracy of endometrial biopsy data in molecular classification remains uncertain. This study aimed to assess the concordance of molecular classifications between preoperative biopsy and hysterectomy to predict prognosis before surgical staging. METHODS: Endometrial biopsies and corresponding hysterectomy specimens were collected at the National Cancer Center Hospital between 2012 and 2023. Immunohistochemistry for p53 and mismatch repair (MMR) proteins and next-generation sequencing of all exons of polymerase epsilon (POLE) were performed. Given the limited number of POLE mut cases in prior studies, we prepared a POLE mut-enriched cohort. Cohen's kappa estimates were used to determine concordance for molecular and clinicopathological subgroup assignments. RESULTS: Among 70 patients classified into four molecular subtype groups, 33 exhibited POLE mutations, 15 showed loss of MMR protein expression, 13 had p53-abnormality, and 9 had no specific molecular profile. Concordance between biopsy and hysterectomy specimens was 100% (κ = 1.000). In contrast, histological types and grades between biopsy and surgical specimens showed moderate and substantial agreement (κ = 0.420 and κ = 0.780, respectively). CONCLUSIONS: Molecular subtypes were completely consistent with those derived from surgical specimens, demonstrating high concordance between preoperative and postoperative molecular classifications. This suggests that endometrial biopsies could reliably predict prognosis. Future studies should investigate how biopsy-based molecular profiling influences treatment planning and patient outcomes.

Genomic profiles of Japanese patients with vulvar squamous cell carcinoma.

The incidence of vulvar carcinoma varies by race; however, it is a rare disease, and its genomic profiles remain largely unknown. This study examined the characteristics of vulvar squamous cell carcinoma (VSCC) in Japanese patients, focusing on genomic profiles and potential racial disparities. The study included two Japanese groups: the National Cancer Center Hospital (NCCH) group comprised 19 patients diagnosed between 2015 and 2023, and the Center for Cancer Genomics and Advanced Therapeutics group comprised 29 patients diagnosed between 2019 and 2022. Somatic mutations were identified by targeted or panel sequencing, and TP53 was identified as the most common mutation (52-81%), followed by HRAS (7-26%), CDKN2A (21-24%), and PIK3CA (5-10%). The mutation frequencies, except for TP53, were similar to those of Caucasian cohorts. In the NCCH group, 16 patients of HPV-independent tumors were identified by immunohistochemistry and genotyping. Univariate analysis revealed that TP53-mutated patients were associated with a poor prognosis (log-rank test, P = 0.089). Japanese VSCC mutations resembled those of Caucasian vulvar carcinomas, and TP53 mutations predicted prognosis regardless of ethnicity. The present findings suggest potential molecular-targeted therapies for select VSCC patients.

Mechanism of ERBB2 gene overexpression by the formation of super-enhancer with genomic structural abnormalities in lung adenocarcinoma without clinically actionable genetic alterations.

BACKGROUND: In an extensive genomic analysis of lung adenocarcinomas (LUADs), driver mutations have been recognized as potential targets for molecular therapy. However, there remain cases where target genes are not identified. Super-enhancers and structural variants are frequently identified in several hundred loci per case. Despite this, most cancer research has approached the analysis of these data sets separately, without merging and comparing the data, and there are no examples of integrated analysis in LUAD. METHODS: We performed an integrated analysis of super-enhancers and structural variants in a cohort of 174 LUAD cases that lacked clinically actionable genetic alterations. To achieve this, we conducted both WGS and H3K27Ac ChIP-seq analyses using samples with driver gene mutations and those without, allowing for a comprehensive investigation of the potential roles of super-enhancer in LUAD cases. RESULTS: We demonstrate that most genes situated in these overlapped regions were associated with known and previously unknown driver genes and aberrant expression resulting from the formation of super-enhancers accompanied by genomic structural abnormalities. Hi-C and long-read sequencing data further corroborated this insight. When we employed CRISPR-Cas9 to induce structural abnormalities that mimicked cases with outlier ERBB2 gene expression, we observed an elevation in ERBB2 expression. These abnormalities are associated with a higher risk of recurrence after surgery, irrespective of the presence or absence of driver mutations. CONCLUSIONS: Our findings suggest that aberrant gene expression linked to structural polymorphisms can significantly impact personalized cancer treatment by facilitating the identification of driver mutations and prognostic factors, contributing to a more comprehensive understanding of LUAD pathogenesis.

International consensus guidelines for the definition, detection, and interpretation of autophagy-dependent ferroptosis.

Macroautophagy/autophagy is a complex degradation process with a dual role in cell death that is influenced by the cell types that are involved and the stressors they are exposed to. Ferroptosis is an iron-dependent oxidative form of cell death characterized by unrestricted lipid peroxidation in the context of heterogeneous and plastic mechanisms. Recent studies have shed light on the involvement of specific types of autophagy (e.g. ferritinophagy, lipophagy, and clockophagy) in initiating or executing ferroptotic cell death through the selective degradation of anti-injury proteins or organelles. Conversely, other forms of selective autophagy (e.g. reticulophagy and lysophagy) enhance the cellular defense against ferroptotic damage. Dysregulated autophagy-dependent ferroptosis has implications for a diverse range of pathological conditions. This review aims to present an updated definition of autophagy-dependent ferroptosis, discuss influential substrates and receptors, outline experimental methods, and propose guidelines for interpreting the results.Abbreviation: 3-MA:3-methyladenine; 4HNE: 4-hydroxynonenal; ACD: accidentalcell death; ADF: autophagy-dependentferroptosis; ARE: antioxidant response element; BH2:dihydrobiopterin; BH4: tetrahydrobiopterin; BMDMs: bonemarrow-derived macrophages; CMA: chaperone-mediated autophagy; CQ:chloroquine; DAMPs: danger/damage-associated molecular patterns; EMT,epithelial-mesenchymal transition; EPR: electronparamagnetic resonance; ER, endoplasmic reticulum; FRET: Försterresonance energy transfer; GFP: green fluorescent protein;GSH: glutathione;IF: immunofluorescence; IHC: immunohistochemistry; IOP, intraocularpressure; IRI: ischemia-reperfusion injury; LAA: linoleamide alkyne;MDA: malondialdehyde; PGSK: Phen Green™ SK;RCD: regulatedcell death; PUFAs: polyunsaturated fatty acids; RFP: red fluorescentprotein;ROS: reactive oxygen species; TBA: thiobarbituricacid; TBARS: thiobarbituric acid reactive substances; TEM:transmission electron microscopy.

Clinical features and impact of p53 status on sporadic mismatch repair deficiency and Lynch syndrome in uterine cancer.

The clinical features of sporadic mismatch repair deficiency (MMRd) and Lynch syndrome (LS) in Japanese patients with endometrial cancer (EC) were examined by evaluating the prevalence and prognostic factors of LS and sporadic MMRd in patients with EC. Targeted sequencing of five LS susceptibility genes (MLH1, MSH2, MSH6, PMS2, and EPCAM) was carried out in 443 patients with EC who were pathologically diagnosed with EC at the National Cancer Center Hospital between 2011 and 2018. Pathogenic variants in these genes were detected in 16 patients (3.7%). Immunohistochemistry for MMR proteins was undertaken in 337 of the 433 (77.9%) EC patients, and 91 patients (27.0%) showed absent expression of at least one MMR protein. The 13 cases of LS with MMR protein loss (93.8%) showed a favorable prognosis with a 5-year overall survival (OS) rate of 100%, although there was no statistically significant difference between this group and the sporadic MMRd group (p = 0.27). In the MMRd without LS group, the 5-year OS rate was significantly worse in seven patients with an aberrant p53 expression pattern than in those with p53 WT (53.6% vs. 93.9%, log-rank test; p = 0.0016). These results suggest that p53 abnormalities and pathogenic germline variants in MMR genes could be potential biomarkers for the molecular classification of EC with MMRd.

Advances in cancer DNA methylation analysis with methPLIER: use of non-negative matrix factorization and knowledge-based constraints to enhance biological interpretability.

DNA methylation is an epigenetic modification that results in dynamic changes during ontogenesis and cell differentiation. DNA methylation patterns regulate gene expression and have been widely researched. While tools for DNA methylation analysis have been developed, most of them have focused on intergroup comparative analysis within a dataset; therefore, it is difficult to conduct cross-dataset studies, such as rare disease studies or cross-institutional studies. This study describes a novel method for DNA methylation analysis, namely, methPLIER, which enables interdataset comparative analyses. methPLIER combines Pathway Level Information Extractor (PLIER), which is a non-negative matrix factorization (NMF) method, with regularization by a knowledge matrix and transfer learning. methPLIER can be used to perform intersample and interdataset comparative analysis based on latent feature matrices, which are obtained via matrix factorization of large-scale data, and factor-loading matrices, which are obtained through matrix factorization of the data to be analyzed. We used methPLIER to analyze a lung cancer dataset and confirmed that the data decomposition reflected sample characteristics for recurrence-free survival. Moreover, methPLIER can analyze data obtained via different preprocessing methods, thereby reducing distributional bias among datasets due to preprocessing. Furthermore, methPLIER can be employed for comparative analyses of methylation data obtained from different platforms, thereby reducing bias in data distribution due to platform differences. methPLIER is expected to facilitate cross-sectional DNA methylation data analysis and enhance DNA methylation data resources.

Vault-phagy: a phase-separation-mediated selective autophagy of vault, a non-membranous organelle.

SQSTM1/p62 bodies are phase-separated condensates that play a fundamental role in intracellular quality control and stress responses. Despite extensive studies investigating the mechanism of formation and degradation of SQSTM1/p62 bodies, the constituents of SQSTM1/p62 bodies remain elusive. We recently developed a purification method for intracellular SQSTM1/p62 bodies using a cell sorter and identified their constituents by mass spectrometry. Combined with mass spectrometry of tissues from selective autophagy-deficient mice, we identified vault, a ubiquitous non-membranous organelle composed of proteins and non-coding RNA, as a novel substrate for selective autophagy. Vault directly binds to NBR1, an SQSTM1/p62 binding partner recruited to SQSTM1/p62 bodies, and is subsequently degraded by selective autophagy dependent on the phase separation of SQSTM1/p62. We named this process "vault-phagy" and found that defects in vault-phagy are related to nonalcoholic steatohepatitis (NASH)-derived hepatocellular carcinoma. Our method for purifying SQSTM1/p62 bodies will contribute to elucidating the mechanisms of several stress responses and diseases mediated by SQSTM1/p62 bodies.

p62 bodies: cytosolic zoning by phase separation.

Cellular zoning or partitioning is critical in preventing macromolecules from random diffusion and orchestrating the spatiotemporal dynamics of biochemical reactions. Along with membranous organelles, membraneless organelles contribute to the precise regulation of biochemical reactions inside cells. In response to environmental cues, membraneless organelles rapidly form through liquid-liquid phase separation, sequester certain proteins and RNAs, mediate specific reactions and dissociate. Among membraneless organelles, ubiquitin-positive condensates, namely, p62 bodies, maintain cellular homeostasis through selective autophagy of themselves to contribute to intracellular quality control. p62 bodies also activate the anti-oxidative stress response regulated by the KEAP1-NRF2 system. In this review, we present an overview of recent advancements in cellular and molecular biology related to p62 bodies, highlighting their dynamic nature and functions.

The UFM1 system: Working principles, cellular functions, and pathophysiology.

Ubiquitin-fold modifier 1 (UFM1) is a ubiquitin-like protein covalently conjugated with intracellular proteins through UFMylation, a process similar to ubiquitylation. Growing lines of evidence regarding not only the structural basis of the components essential for UFMylation but also their biological properties shed light on crucial roles of the UFM1 system in the endoplasmic reticulum (ER), such as ER-phagy and ribosome-associated quality control at the ER, although there are some functions unrelated to the ER. Mouse genetics studies also revealed the indispensable roles of this system in hematopoiesis, liver development, neurogenesis, and chondrogenesis. Of critical importance, mutations of genes encoding core components of the UFM1 system in humans cause hereditary developmental epileptic encephalopathy and Schohat-type osteochondrodysplasia of the epiphysis. Here, we provide a multidisciplinary review of our current understanding of the mechanisms and cellular functions of the UFM1 system as well as its pathophysiological roles, and discuss issues that require resolution.

Integrative analysis reveals early epigenetic alterations in high-grade serous ovarian carcinomas.

High-grade serous ovarian carcinoma (HGSOC) is the most lethal gynecological malignancy. To date, the profiles of gene mutations and copy number alterations in HGSOC have been well characterized. However, the patterns of epigenetic alterations and transcription factor dysregulation in HGSOC have not yet been fully elucidated. In this study, we performed integrative omics analyses of a series of stepwise HGSOC model cells originating from human fallopian tube secretory epithelial cells (HFTSECs) to investigate early epigenetic alterations in HGSOC tumorigenesis. Assay for transposase-accessible chromatin using sequencing (ATAC-seq), chromatin immunoprecipitation sequencing (ChIP-seq), and RNA sequencing (RNA-seq) methods were used to analyze HGSOC samples. Additionally, protein expression changes in target genes were confirmed using normal HFTSECs, serous tubal intraepithelial carcinomas (STICs), and HGSOC tissues. Transcription factor motif analysis revealed that the DNA-binding activity of the AP-1 complex and GATA family proteins was dysregulated during early tumorigenesis. The protein expression levels of JUN and FOSL2 were increased, and those of GATA6 and DAB2 were decreased in STIC lesions, which were associated with epithelial-mesenchymal transition (EMT) and proteasome downregulation. The genomic region around the FRA16D site, containing a cadherin cluster region, was epigenetically suppressed by oncogenic signaling. Proteasome inhibition caused the upregulation of chemokine genes, which may facilitate immune evasion during HGSOC tumorigenesis. Importantly, MEK inhibitor treatment reversed these oncogenic alterations, indicating its clinical effectiveness in a subgroup of patients with HGSOC. This result suggests that MEK inhibitor therapy may be an effective treatment option for chemotherapy-resistant HGSOC.

ATG9A supports Chlamydia trachomatis infection via autophagy-independent mechanisms.

Chlamydia trachomatis infection can be regulated by autophagy-related (ATG) genes. Here, we found that the depletion of ATG9A, one of the core ATG genes, in HeLa cells suppressed C. trachomatis growth in the inclusion. The growth was restored by re-expressing ATG9A or an ATG9A mutant impairing lipid scramblase activity in ATG9A-knockout (KO) cells. Moreover, the depletion of lipid transfer proteins ATG2A/B, responsible for isolation membrane expansion together with ATG9A, did not significantly alter the growth, suggesting that the non-autophagic function of ATG9A supports C. trachomatis infection. ATG9A-KO cells showed no infection-induced redistribution of the Golgi from the perinuclear region to inclusion, which was restored by re-expressing the mutant but not the ATG9A mutant lacking an N-terminal adapter protein-binding domain. Re-expression of the N-terminal deletion mutant in ATG9A-KO cells did not rescue C. trachomatis growth, suggesting the importance of this domain for its growth. Although ATG9A-KO cells showed enhanced TBK1 activation, interferon (IFN)-ß was not significantly increased, excluding the possibility that upregulation of stimulator of IFN genes (STING) signaling suppressed bacterial growth. Taken together, these findings suggest that the proper trafficking, rather than the isolation membrane expansion function, of ATG9A assists C. trachomatis growth in the inclusion. IMPORTANCE ATG9A is an autophagy-related gene that functions during the isolation membrane expansion process to form autophagosomes, but it also has other functions independent of autophagy. In this study, we employed ATG9A-deficient HeLa cells and found that the absence of ATG9A negatively impacted proliferation of Chlamydia trachomatis in inclusions. Furthermore, rescue experiments using ATG9A mutants revealed that this action was mediated not by its autophagic function but by its binding ability to clathrin adapter proteins. These findings suggest that the proper trafficking of ATG9A assists C. trachomatis growth in the inclusion.

Rare FGFR fusion genes in cervical cancer and transcriptome-based subgrouping of patients with a poor prognosis.

BACKGROUND: Although cervical cancer is often characterized as preventable, its incidence continues to increase in low- and middle-income countries, underscoring the need to develop novel therapeutics for this disease.This study assessed the distribution of fusion genes across cancer types and used an RNA-based classification to divide cervical cancer patients with a poor prognosis into subgroups. MATERIAL AND METHODS: RNA sequencing of 116 patients with cervical cancer was conducted. Fusion genes were extracted using StarFusion program. To identify a high-risk group for recurrence, 65 patients who received postoperative adjuvant therapy were subjected to non-negative matrix factorization to identify differentially expressed genes between recurrent and nonrecurrent groups. RESULTS: We identified three cases with FGFR3-TACC3 and one with GOPC-ROS1 fusion genes as potential targets. A search of publicly available data from cBioPortal (21,789 cases) and the Center for Cancer Genomics and Advanced Therapeutics (32,608 cases) showed that the FGFR3 fusion is present in 1.5% and 0.6% of patients with cervical cancer, respectively. The frequency of the FGFR3 fusion gene was higher in cervical cancer than in other cancers, regardless of ethnicity. Non-negative matrix factorization identified that the patients were classified into four Basis groups. Pathway enrichment analysis identified more extracellular matrix kinetics dysregulation in Basis 3 and more immune system dysregulation in Basis 4 than in the good prognosis group. CIBERSORT analysis showed that the fraction of M1 macrophages was lower in the poor prognosis group than in the good prognosis group. CONCLUSIONS: The distribution of FGFR fusion genes in patients with cervical cancer was determined by RNA-based analysis and used to classify patients into clinically relevant subgroups.

Mechanistic insights into the roles of the UFM1 E3 ligase complex in ufmylation and ribosome-associated protein quality control.

Ubiquitin-fold modifier 1 (UFM1) is a ubiquitin-like protein covalently conjugated with intracellular proteins through ufmylation, similar to ubiquitylation. Ufmylation is involved in processes such as endoplasmic reticulum (ER)-associated protein degradation, ribosome-associated protein quality control (RQC) at the ER (ER-RQC), and ER-phagy. However, it remains unclear how ufmylation regulates such distinct ER-related functions. Here, we provide insights into the mechanism of the UFM1 E3 complex in not only ufmylation but also ER-RQC. The E3 complex consisting of UFL1 and UFBP1 interacted with UFC1, UFM1 E2, and, subsequently, CDK5RAP3, an adaptor for ufmylation of ribosomal subunit RPL26. Upon disome formation, the E3 complex associated with ufmylated RPL26 on the 60S subunit through the UFM1-interacting region of UFBP1. Loss of E3 components or disruption of the interaction between UFBP1 and ufmylated RPL26 attenuated ER-RQC. These results provide insights into not only the molecular basis of the ufmylation but also its role in proteostasis.

Phosphorylation of phase-separated p62 bodies by ULK1 activates a redox-independent stress response.

NRF2 is a transcription factor responsible for antioxidant stress responses that is usually regulated in a redox-dependent manner. p62 bodies formed by liquid-liquid phase separation contain Ser349-phosphorylated p62, which participates in the redox-independent activation of NRF2. However, the regulatory mechanism and physiological significance of p62 phosphorylation remain unclear. Here, we identify ULK1 as a kinase responsible for the phosphorylation of p62. ULK1 colocalizes with p62 bodies, directly interacting with p62. ULK1-dependent phosphorylation of p62 allows KEAP1 to be retained within p62 bodies, thus activating NRF2. p62S351E/+ mice are phosphomimetic knock-in mice in which Ser351, corresponding to human Ser349, is replaced by Glu. These mice, but not their phosphodefective p62S351A/S351A counterparts, exhibit NRF2 hyperactivation and growth retardation. This retardation is caused by malnutrition and dehydration due to obstruction of the esophagus and forestomach secondary to hyperkeratosis, a phenotype also observed in systemic Keap1-knockout mice. Our results expand our understanding of the physiological importance of the redox-independent NRF2 activation pathway and provide new insights into the role of phase separation in this process.

Integrated proteomics identifies p62-dependent selective autophagy of the supramolecular vault complex.

In addition to membranous organelles, autophagy selectively degrades biomolecular condensates, in particular p62/SQSTM1 bodies, to prevent diseases including cancer. Evidence is growing regarding the mechanisms by which autophagy degrades p62 bodies, but little is known about their constituents. Here, we established a fluorescence-activated-particle-sorting-based purification method for p62 bodies using human cell lines and determined their constituents by mass spectrometry. Combined with mass spectrometry of selective-autophagy-defective mouse tissues, we identified vault, a large supramolecular complex, as a cargo within p62 bodies. Mechanistically, major vault protein directly interacts with NBR1, a p62-interacting protein, to recruit vault into p62 bodies for efficient degradation. This process, named vault-phagy, regulates homeostatic vault levels in vivo, and its impairment may be associated with non-alcoholic-steatohepatitis-derived hepatocellular carcinoma. Our study provides an approach to identifying phase-separation-mediated selective autophagy cargoes, expanding our understanding of the role of phase separation in proteostasis.

Analysis of super-enhancer using machine learning and its application to medical biology.

The analysis of super-enhancers (SEs) has recently attracted attention in elucidating the molecular mechanisms of cancer and other diseases. SEs are genomic structures that strongly induce gene expression and have been reported to contribute to the overexpression of oncogenes. Because the analysis of SEs and integrated analysis with other data are performed using large amounts of genome-wide data, artificial intelligence technology, with machine learning at its core, has recently begun to be utilized. In promoting precision medicine, it is important to consider information from SEs in addition to genomic data; therefore, machine learning technology is expected to be introduced appropriately in terms of building a robust analysis platform with a high generalization performance. In this review, we explain the history and principles of SE, and the results of SE analysis using state-of-the-art machine learning and integrated analysis with other data are presented to provide a comprehensive understanding of the current status of SE analysis in the field of medical biology. Additionally, we compared the accuracy between existing machine learning methods on the benchmark dataset and attempted to explore the kind of data preprocessing and integration work needed to make the existing algorithms work on the benchmark dataset. Furthermore, we discuss the issues and future directions of current SE analysis.

Utility of molecular subtypes and genetic alterations for evaluating clinical outcomes in 1029 patients with endometrial cancer.

BACKGROUND: We investigated the utility of a molecular classifier tool and genetic alterations for predicting prognosis in Japanese patients with endometrial cancer. METHODS: A total of 1029 patients with endometrial cancer from two independent cohorts were classified into four molecular subtype groups. The primary and secondary endpoints were relapse-free survival (RFS) and overall survival (OS), respectively. RESULTS: Among the 265 patients who underwent initial surgery, classified according to immunohistochemistry, patients with DNA polymerase epsilon exonuclease domain mutation had an excellent prognosis (RFS and OS), patients with no specific molecular profile (NSMP) and mismatch repair protein deficiency had an intermediate prognosis, and those with protein 53 abnormal expression (p53abn) had the worst prognosis (P < 0.001). In the NSMP group, mutant KRAS and wild-type ARID1A were associated with significantly poorer 5-year RFS (41.2%) than other genomic characteristics (P < 0.001). The distribution of the subtypes differed significantly between patients with recurrence/progression and classified by sequencing (n = 764) and patients who underwent initial surgery (P < 0.001). Among patients with recurrence/progression, 51.4% had the opportunity to receive molecular targeted therapy. CONCLUSIONS: A molecular classifier is a useful tool for determining prognosis and eligibility for molecularly targeted therapy in patients with endometrial cancer.