Cancer Data Research Articles

Abstract PR003: Structure-guided design and optimization of small molecule CD73 inhibitors with excellent drug-like properties: discovery of quemliclustat

Abstract In the tumor microenvironment, high concentrations of extracellular adenosine promote tumor proliferation through various immunosuppressive mechanisms. High expression of CD73—an ecto-nucleotidase that produces adenosine from adenosine monophosphate—has been associated with poorer prognosis in several tumor types. Herein, we describe the drug discovery efforts that resulted in the discovery of quemliclustat, a highly potent and selective inhibitor of CD73. As a small molecule with excellent drug-like properties, quemliclustat offers several advantages over CD73 antibodies in development, such as greater inhibition of CD73 enzymatic activity (both soluble and membrane-bound), deeper tumor penetration, and the potential for both IV and oral formulations. Quemliclustat was the first small molecule CD73 inhibitor to enter clinical development and is currently being evaluated in various Phase 1/2 studies in advanced solid tumors, including lung and pancreatic cancers. Recent data in metastatic pancreatic cancer are supportive of further study in this disease setting. Citation Format: Jenna Jeffrey Structure-guided design and optimization of small molecule CD73 inhibitors with excellent drug-like properties: discovery of quemliclustat. [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Optimizing Therapeutic Efficacy and Tolerability through Cancer Chemistry; 2024 Dec 9-11; Toronto, Ontario, Canada. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(12_Suppl):Abstract nr PR003

A Prostate Imaging-Reporting and Data System version 2.1-based predictive model for clinically significant prostate cancer diagnosis.

To develop and validate a Prostate Imaging-Reporting and Data System (PI-RADS) version 2.1 (v2.1)-based predictive model for diagnosis of clinically significant prostate cancer (csPCa), integrating clinical and multiparametric magnetic resonance imaging (mpMRI) data, and compare its performance with existing models. We retrospectively analysed data from patients who underwent prospective mpMRI assessment using the PI-RADS v2.1 scoring system and biopsy at our institution between April 2019 and December 2023. A 'Clinical Baseline' model using patient demographics and laboratory results and an 'MRI Added' model additionally incorporating PI-RADS v2.1 scores and prostate volumes were created and validated on internal and external patients. Both models were compared against two previously published MRI-based algorithms for csPCa using area under the receiver operating characteristic curve (AUC) and decision curve analysis. A total of 1319 patients across internal and external cohorts were included. Our 'MRI Added' model demonstrated significantly improved discriminative ability (AUCinternal 0.88, AUCexternal 0.79) compared to our 'Clinical Baseline' model (AUCinternal 0.75, AUCexternal 0.68) (P < 0.001). The 'MRI Added' model also showed higher net benefits across various clinical threshold probabilities and compared to a 'biopsy all' approach, it reduced unnecessary biopsies (defined as biopsies without Gleason Grade Group ≥2 csPCa) by 27% in the internal cohort and 10% in the external cohort at a risk threshold of 25%. However, there was no significant difference in predictive ability and reduction in unnecessary biopsies between our model and comparative ones developed for PI-RADS v2 and v1. Our PI-RADS v2.1-based mpMRI model significantly enhances csPCa prediction, outperforming the traditional clinical model in accuracy and reduction of unnecessary biopsies. It proves promising across diverse patient populations, establishing an updated, integrated approach for detection and management of prostate cancer.

Clinical Characteristics, Management and Outcomes of Colitis-associated Colorectal Cancer and the Comparison to Sporadic Colorectal Cancer in Taiwan.

We explored the clinical characteristics, treatment, and outcomes of colitis-associated colorectal cancer and compared with sporadic colorectal cancer in Taiwan. In this retrospective study spanning 1987-2022, colitis-associated colorectal cancers diagnosed according to endoscopic and pathological reports from 14 tertiary centers were reported to our cohort. Clinical demographics, endoscopic findings, histological results, treatment modalities, and outcomes were analyzed. Sporadic colorectal cancer data were retrieved from the Cancer Registry Annual Report, Ministry of Health and Welfare, Taiwan. We enrolled 65 colitis-associated colorectal cancer patients (median age: 56 years; male: 66.2%). Distal colon was the most common tumor location (41.5%). Of ulcerative colitis patients, 77.2% had extensive colitis and 76.5% had Mayo endoscopic subscores of ≥2. Moreover, 50% of lesions were nonpolypoid with indistinct borders in 66.7%. Signet-ring cell subtype consisted 12.3%. Surveillance colonoscopy adherence was 78.4%, yet 51.3% interval cancers occurred. Disease stage 0-4 distribution was 15%, 20%, 13.3%, 20%, and 31.7%, respectively. Endoscopic resection was feasible for 14%, while 67.7% required surgery. During follow-up (median: 21.5 months), we recorded 23.2% recurrence and 34.5% mortality. Lesions with indistinct borders were associated with adverse outcomes (adjusted odds ratio = 11.5 [1.35-98.16]). Colitis-associated rectal cancers, diagnosed later (p < 0.001), had worse outcomes than sporadic rectal cancers. This is the largest Asian colitis-associated colorectal cancer cohort study, emphasizing the need for stringent disease control, improving detection and reducing interval cancers. Signet-ring cell subtype was prevalent. Rectal colitis-associated cancers were diagnosed later with poorer outcomes then sporadic rectal cancers.

Cardiometabolic risk factors among thyroid cancer survivors

IntroductionThyroid cancer incidence has tripled in the last decades, resulting in a growing population of survivors. In addition to the risks associated with the cancer itself, cardiovascular disease represents a major cause of morbidity and mortality in thyroid cancer survivors. Our study aims to assess the frequency of cardiovascular and metabolic risk factors in thyroid cancer survivors (TCS).MethodsA cross-sectional analysis of thyroid cancer patients at Constantine’s University Hospital, Algeria (January 2021 to July 2022), used the WHO STEPS approach to assess risk factors. Cancer data were extracted from medical records.ResultsA total of 342 patients were included, of whom 88.9% were women. The mean age was 48 years (range 21–85), papillary thyroid cancer accounted for 82% of cases. The mean survival duration was 4 years (6 months to 32 years), and 74.5% of patients responded well to initial treatment. The current smoking rate was 2.6%, 7.5% of patients engaged in vigorous physical activity, and only 7.8% consumed more than 5 fruits or vegetables daily. At the study time, over 80.7% were overweight or obese, and 86% exhibited android obesity. Hypertension (48.8%), diabetes (27.5%), prediabetes (32.5%), dyslipidemia (77.5%), and metabolic syndrome (61.1%) were common. Factors associated with metabolic syndrome included a family history of obesity (OR) of 1.80 (1.08–3.02), age over 40 OR of 2.71 (1.63–4.51), overweight/obesity OR of 5.134 (2.68–9.84), and levothyroxine doses ≥ 125 µg/day OR of 2.480 (1.45–4.23).ConclusionCardiovascular and metabolic risk factors are prevalent in our population of thyroid cancer survivors.

A Novel Sensitivity Maximization at a Given Specificity Method for Binary Classifications

Abstract In the cancer early detection field, logistic regression is a frequently used approach to establish a combination rule that differentiates cancer from non-cancer. However, the application of logistic regression relies on a maximum likelihood approach, which may not yield optimal combination rules for maximizing sensitivity at a clinically desirable specificity and vice versa. Here, we have developed an improved regression framework, Sensitivity Maximization At a Given Specificity, SMAGS, for binary classification that finds the linear decision rule yielding the maximum sensitivity for a given specificity or the maximum specificity for a given sensitivity. We additionally expand the framework for feature selection that satisfies sensitivity and specificity maximizations. We compare our SMAGS method with normal logistic regression using two synthetic datasets and reported data for colorectal cancer (CRC) from the 2018 CancerSEEK study. In the CRC CancerSEEK dataset, we report 14% improvement in sensitivity at 98.5% specificity (0.31 vs 0.57; p-value:&amp;lt;0.05). The SMAGS method provides an alternative to logistic regression for modeling combination rules for biomarkers and early detection applications.

Novel Systemic Anticancer Therapy and Healthcare Utilization at the End of Life: A Retrospective Cohort Study.

Novel systemic anticancer therapies (SACT) in the form of targeted and immunotherapies are increasingly replacing traditional chemotherapy. Little is known about the impact of novel SACT on healthcare resource utilization (HCRU) at the end of life. A retrospective review of patients attending a tertiary cancer center in Toronto, Canada, with advanced solid or hematological malignancies who died in 2019. Demographic and cancer data, SACT use, HCRU (emergency room [ER] visits, acute/intensive care unit [ICU] admission, and place of death) were retrieved and compared between those who received SACT in their last 30 days of life and those who did not. Chi-squared tests or Quasi-Poisson regression calculated HCRU expressed as percentages or rate ratios (RR). Univariate and multivariable logistic regression identified factors independently associated with SACT use. Of 443 patients included, 88 (20%) received SACT in the last 30 days of life, with 42 (48%) receiving targeted therapies and 10 (11%) immunotherapy. Factors associated with SACT use included younger age (p = 0.016), breast (p < 0.001), lung (p = 0.047), hematological malignancies (p = 0.002), fewer comorbidities (p = 0.039), and novel SACT (p = 0.006). Receipt of SACT was associated with a higher frequency of ER visits (55% vs. 36% who did not receive SACT, p = 0.001), acute hospitalizations (68% vs. 47%, p < 0.001), ICU admissions (18% vs. 7%, p = 0.003), and death in hospital (45% vs. 30%, p = 0.008). Novel SACT use at the end of life is high and is strongly associated with HCRU. Future studies should explore the impact of advance care planning and palliative care referrals on SACT use.

Revealing key proteins in comparison of tumor and para-tumor tissues in stage I esophageal squamous cell carcinoma: A combined gene expression clustering and protein interaction network analysis

Esophageal Squamous Cell Carcinoma (ESCC) which is diagnosed in advanced stages and metastasis to the vital organs, remained the sixth fatal malignancy among other types of cancer-related death. Considering the lack of specific clinical symptoms in the early development of the disease, finding suitable biomarkers and tissue-specific alterations for effective screening of ESCC is targeted in this project. Gene expression profiles of tumor tissue and para-tumors of stage I ESCC patients were retrieved from the Gene Expression Omnibus (GEO) database. The significant tissue-specific differentially expressed genes (DEGs) were identified and studied based on fold change distribution and the functional role of genes in the action map. A total number of 11,483 significant DEGs discriminated the tumor tissue from para-tumor tissue. Clustering analysis of significant DEGs led to identifying 220 DEGs as the final significant genes. Protein interaction network and action map analysis of the final gene list showed matrix metallopeptidase-9 (MMP9) as the critical gene related to the development of ESCC diseases in stage I. Significant expression change of MMP9 in esophageal carcinoma was validated in UALCAN (The University of Alabama at Birmingham Cancer Data Analysis Portal). This study highlighted the pivotal role of MMP9 in combination with SPP1, MMP13, and IL18 as a possible biomarkers panel in studies of tumor invasion and prognosis for stage I ESCC disease.

A clustering approach to integrative analyses of multiomic cancer data

Rapid technological advances have allowed for molecular profiling across multiple omics domains for clinical decision-making in many diseases, especially cancer. However, as tumor development and progression are biological processes involving composite genomic aberrations, key challenges are to effectively assimilate information from these domains to identify genomic signatures and druggable biological entities, develop accurate risk prediction profiles for future patients, and identify novel patient subgroups for tailored therapy and monitoring. We propose integrative frameworks for high-dimensional multiple-domain cancer data. These Bayesian mixture model-based approaches coherently incorporate dependence within and between domains to accurately detect tumor subtypes, thus providing a catalog of genomic aberrations associated with cancer taxonomy. The flexible and scalable Bayesian nonparametric strategy performs simultaneous bidirectional clustering of the tumor samples and genomic probes to achieve dimension reduction. We describe an efficient variable selection procedure that can identify relevant genomic aberrations and potentially reveal underlying drivers of disease. Although the work is motivated by lung cancer datasets, the proposed methods are broadly applicable in a variety of contexts involving high-dimensional data. The success of the methodology is demonstrated using artificial data and lung cancer omics profiles publicly available from The Cancer Genome Atlas.

The risk of malignancy in patients with psoriasis treated with long-term tumor necrosis factor-α inhibitor: a systematic review and meta-analysis.

The relationship between tumor necrosis factor-α inhibitors (TNFi) and cancer risk is complex, given their pivotal role in managing chronic inflammatory conditions. Persistent concerns about TNFi therapy potentially increasing cancer risk necessitate a thorough understanding of its long-term effects on cancer development in patients with psoriasis to guide therapeutic decision-making. This systematic review and meta-analysis aimed to evaluate the association between long-term TNFi therapy and cancer risk in patients with psoriasis. Data were collected from PubMed, Embase, and the Cochrane Library, from their inception to January 1, 2024. The studies included adults with psoriasis or psoriatic arthritis receiving TNFi, presenting standardized incidence ratio (SIR) data for cancer. Data extraction followed PRISMA guidelines, with independent extraction by two authors, and pooled data synthesis was conducted using a random-effects model. The primary outcomes were SIRs for all cancers excluding non-melanoma skin carcinoma (NMSC) and for NMSC itself. The secondary outcome was the SIR of specific cancer types. The meta-analysis included eight studies with a total of 32,765 psoriasis patients. The pooled results showed no increased risk of cancers, including all cancers excluding NMSC, melanoma, lymphoma, prostate cancer, and breast cancer, among individuals receiving long-term TNFi therapy for psoriasis compared to the general population. However, subgroup analysis found an elevated risk of NMSC in patients with psoriatic arthritis and a significant increase in the risk of squamous cell carcinoma (SCC) among psoriasis patients treated with TNFi compared to the general population (SIR, 1.84; 95% CI, 1.16 - 2.92 and SIR, 2.84; 95% CI, 1.64 - 4.91, respectively). Our systematic review and meta-analysis indicate that most cancers examined did not demonstrate an increased risk following long-term TNFi therapy in psoriasis patients. However, for patients with psoriatic arthritis or those at high risk of SCC, these findings underscore the importance of personalized treatment strategies that weigh individual risks and benefits.

C/EBPβ-dependent autophagy inhibition hinders NK cell function in cancer

NK cells are endowed with tumor killing ability, nevertheless most cancers impair NK cell functionality, and cell-based therapies have limited efficacy in solid tumors. How cancers render NK cell dysfunctional is unclear, and overcoming resistance is an important immune-therapeutic aim. Here, we identify autophagy as a central regulator of NK cell anti-tumor function. Analysis of differentially expressed genes in tumor-infiltrating versus non-tumor NK cells from our previously published scRNA-seq data of advanced human prostate cancer shows deregulation of the autophagic pathway in tumor-infiltrating NK cells. We confirm this by flow cytometry in patients and in diverse cancer models in mice. We further demonstrate that exposure of NK cells to cancer deregulates the autophagic process, decreases mitochondrial polarization and impairs effector functions. Mechanistically, CCAAT enhancer binding protein beta (C/EBPβ), downstream of CXCL12-CXCR4 interaction, acts as regulator of NK cell metabolism. Accordingly, inhibition of CXCR4 and C/EBPβ restores NK cell fitness. Finally, genetic and pharmacological activation of autophagy improves NK cell effector and cytotoxic functions, which enables tumour control by NK and CAR-NK cells. In conclusion, our study identifies autophagy as an intracellular checkpoint in NK cells and introduces autophagy regulation as an approach to strengthen NK-cell-based immunotherapies.

Can we empirically derive a geographic definition of ‘coastal’ for use in cancer data reporting? An ecological modelling study using England’s national cancer registry

BackgroundReducing avoidable systematic differences in population health requires first understanding which populations are currently disadvantaged. Although the health of coastal communities in England has been of concern for some years,...

Gradient-induced variable selection in reproducing kernel Hilbert space for survival analysis

The analysis of survival data is often hampered by a potentially very large number of covariates compounded with incomplete data. This paper considers the problem of variable selection where the response is subject to random (right) censoring. We introduce a model-free variable selection procedure via learning the gradients of quantile regression functions with two popular censoring weighting schemes. The key advantage of the proposed approach is that it does not require explicit model assumptions and is computationally efficient. Besides, we also prove its asymptotic consistency and examine its finite sample performance via numerical studies. As an empirical example, the proposed procedure is applied for the analysis of a breast cancer data set.

Type II diabetes and metformin use does not affect colorectal cancer prognosis.

Previous studies on the impact of metformin and colorectal cancer (CRC) outcomes have been limited by small size and confounding by indication, yielding inconsistent results. The aim of this study was to assess whether diabetes and pre-diagnostic metformin use influence CRC prognosis. The study was performed using the Colorectal Cancer Data Base Sweden, a register-linkage originating from the Swedish Colorectal Cancer Register with linkage to national health care registers and demographic registers. All adult patients diagnosed with primary non-metastatic CRC between 2007 and 2016, treated with curative surgery, were identified and followed up from 90 days post-surgery until December 31, 2022. Antidiabetic medication use was defined as dispensed prescription ≥6 months of use within 1 year of surgery. Type II diabetes mellitus (T2DM) patients were divided into three treatment groups (i) diet only, (ii) metformin user, and (iii) non-metformin user. Cox regression models estimated hazard ratios (HRs) with 95% confidence intervals (CIs) for time to recurrence, CRC-specific, and all-cause mortality, adjusted for relevant covariates. Of 33,028 non-metastatic CRC patients, 4539 (13.7%) had T2DM, with 1745 using metformin. A T2DM diagnosis was not associated with increased recurrence rate or CRC-specific mortality; HRadj 0.97 (95% CI 0.89-1.06) and HRadj 0.95 (95% CI 0.87-1.05), respectively, compared with non-diabetic patients. Furthermore, no association between T2DM, metformin use, and recurrence or CRC-specific mortality was seen, HRadj 0.98 (95% CI 0.86-1.12) and HRadj 0.98 (95% CI 0.85-1.13), respectively. T2DM is not associated with an elevated recurrence or CRC-specific mortality. Additionally, metformin use does not impact CRC prognosis.

Identification of a novel prognostic signature for breast cancer derived from post-translational ubiquitin and ubiquitin-like modification-related genes.

Ubiquitin and ubiquitin-like (UUL) modifications play pleiotropic functions and are subject to fine regulatory mechanisms frequently altered in cancer. However, the comprehensive impact of UUL modification on breast cancer remains unclear. Transcriptomic and clinical data of breast cancer were downloaded from TCGA and GEO databases. Molecular subtyping of breast cancer was conducted using the NMF and CIBERSORT algorithms. Prognostic genes were identified via univariate, lasso and multivariate Cox regression analyses. Clinical pathological features, immune cell infiltration, immune therapeutic response and chemotherapy drug sensitivity were compared between groups using the Wilcoxon test. Survival analysis was performed using the Kaplan-Meier method and log-rank test. A total of 63 UUL modification-related genes were differentially expressed, with 29 up-regulated and 34 down-regulated genes. These genes were used to generate two UUL modification patterns that exhibited significant differences in prognostic features and immune cell infiltration. The UUL modification patterns were associated with 2038 differentially expressed genes that were significantly enriched in nuclear division, chromosome segregation, neuroactive ligand-receptor interaction, cell cycle, and other biological processes. Of these genes, 425 were associated with breast cancer prognosis, which enabled the classification of breast cancer into two clusters with significantly distinct prognoses. We developed a prognostic model, UULscore, which comprised nine genes and showed a significant correlation with partial immune cell infiltration. Furthermore, UULscore demonstrated potential predictive value in breast cancer overall survival prediction, immune therapeutic response, and chemotherapy drug sensitivity. UULscore, stage, radiotherapy, and chemotherapy were identified as independent prognostic factors for breast cancer. Based on these factors, a nomogram model was constructed, which demonstrated exceptional prognostic predictive performance. The present study identified two UUL modification-derived molecular subtypes in breast cancer, and have successfully constructed a risk-scoring model that holds potential value in prognosis, immune infiltration, immune therapeutic response, and chemotherapy sensitivity.

MMOSurv: Meta-learning for few-shot survival analysis with multi-omics data.

High-throughput techniques have produced a large amount of high-dimensional multi-omics data, which makes it promising to predict patient survival outcomes more accurately. Recent work has showed the superiority of multi-omics data in survival analysis. However, it remains challenging to integrate multi-omics data to solve few-shot survival prediction problem, with only a few available training samples, especially for rare cancers. In this work, we propose a meta-learning framework for multi-omics few-shot survival analysis, namely MMOSurv, which enables to learn an effective multi-omics survival prediction model from a very few training samples of a specific cancer type, with the meta-knowledge across tasks from relevant cancer types. By assuming a deep Cox survival model with multiple omics, MMOSurv first learns an adaptable parameter initialization for the multi-omics survival model from abundant data of relevant cancers, and then adapts the parameters quickly and efficiently for the target cancer task with a very few training samples. Our experiments on eleven cancer types in TCGA datasets show that, compared to single-omics meta-learning methods, MMOSurv can better utilize the meta-information of similarities and relationships between different omics data from relevant cancer datasets to improve survival prediction of the target cancer with a very few multi-omics training samples. Furthermore, MMOSurv achieves better prediction performance than other state-of-the-art strategies such as multi-task learning and pre-training. MMOSurv is freely available at https://github.com/LiminLi-xjtu/MMOSurv. Supplementary data are available at Bioinformatics online.

Molecular interplay between the upregulated levels of Sad1 and UNC84 Domain Containing 2 (SUN2) and gene expression in medulloblastoma cells.

SUN2 is a nuclear envelope protein associated with the nuclear lamina and with proteins linked to nuclear export, splicing, and nucleo-cytoskeleton communication. Studies of SUN2 in cancer have been limited but have suggested that it has tumor-suppressive activity in some carcinomas. Medulloblastoma is a pediatric tumor that develops in the cerebellum and is currently classified intofour molecular groups: WNT (Wingless), SHH (Sonic Hedgehog), 3, and 4. SUN2 expression profiles appear to be altered in brain cancer but have not been previously evaluated in medulloblastoma. The University of Alabama at Birmingham Cancer (UALCAN) data analysis portal, Gene Expression Profiling Interactive Analysis (GEPIA), the Oncopression gene expression compendium, and the R2 genomics analysis and visualization platform were used to analyze SUN2 expression in cancer, which was found to vary by cancer type; in particular, SUN2 expression was found to be upregulated in medulloblastoma. We also explored the effects of reduced SUN2 protein levels (by RNA interference) on gene expression profiles using a cDNA microarray in DAOY medulloblastoma-derived cells. We found that SUN2 protein is upregulated in medulloblastoma, mainly in the SHH group, which correlates with poor survival. Furthermore, the reduced SUN2 expression in medulloblastoma cells is associated with the downregulation of the expression of other genes, including members of the bitter taste-sensing type 2 receptor (TAS2R) family. This study shows that SUN2 is upregulated in medulloblastoma-with molecular interplay in gene expression-which has group-dependent implications for medulloblastoma development. In particular, the upregulation of SUN2 is associated with a progression of the SHH group of medulloblastoma.

XAI-driven CatBoost multi-layer perceptron neural network for analyzing breast cancer

Early diagnosis of breast cancer is exceptionally important in signifying the treatment results, of women’s health. The present study outlines a novel approach for analyzing breast cancer data by using the CatBoost classification model with a multi-layer perceptron neural network (CatBoost+MLP). Explainable artificial intelligence techniques are used to cohere with the proposed CatBoost with the MLP model. The proposed model aims to enhance the interpretability of predictions in breast cancer diagnosis by leveraging the benefits of CatBoost classification technique in feature identification and also contributing towards the interpretability of the decision model. The proposed CatBoost+MLP has been evaluated using the Shapley additive explanations values to analyze the feature significance in decision-making. Initially, the feature engineering is done using the analysis of variance technique to identify the significant features. The MLP model alone and the CatBoost+MLP model are being analyzed using divergent performance metrics, and the results obtained are compared with contemporary breast cancer identification techniques.

Prediction of Lung Cancer Disease Using Machine Learning Techniques

The pursuit of algorithms utilizing external examples to formulate extensive hypotheses predicting the occurrence of novel instances is recognized, as supervised machine learning (SML). One of the jobs that intelligent systems perform the most frequently is supervised classification. The goal of this work is to evaluate supervised learning algorithms, explain SML classification methodologies, and identify the most effective classification algorithm given the available data. Two distinct machine learning (ML) techniques were examined: Random Forest (RF) and Neural Networks (NN). The algorithms were implemented using Python for knowledge analysis. For the categorization, 310 cases from a lung cancer data set were employed, with 15 features serving as independent variables and one serving as the dependent variable. In comparison to NN classification methods, RF was found to be the algorithm with the highest precision and accuracy, according to the results. The study reveals that while the kappa statistic and mean square error (MSE) are factors on the one hand, the time required to create a model and precision (accuracy) are factors on the other. Consequently, to have supervised predictive ML algorithms need to be precise, accurate, and minimum error. Thus, as a consequence of the research, we are currently at this analysis. The categorizing of NNs accuracy is 0.75 the MSE is 0.25, The RF classification accuracy is 0.89 and the MSE is 0.21.

Research on the mechanism of DTL in the occurrence and development of cancer

Abstract. DTL (denticleless E3 ubiquitin protein ligase homolog) is an E3 ubiquitin ligase that is highly expressed in a variety of tumors and is closely related to the occurrence and development of tumors. Here, we review the latest progress in DTL regulation in various cancer data, including the mechanisms by which changes in its expression affect multiple pathways, ultimately leading to cell cycle arrest and tumor proliferation. Future research should further elucidate the molecular mechanism of DTL and its relationship with tumorigenesis, which is of great significance for the prevention, diagnosis and treatment of tumors. In addition, multi-omics data need to be used to further explore the differential expression and regulatory network of DTL at the single cell level, which is crucial for finding tumor suppressor drug targets.

Generating Biomedical Knowledge Graphs from Knowledge Bases, Registries, and Multiomic Data.

As large clinical and multiomics datasets and knowledge resources accumulate, they need to be transformed into computable and actionable information to support automated reasoning. These datasets range from laboratory experiment results to electronic health records (EHRs). Barriers to accessibility and sharing of such datasets include diversity of content, size and privacy. Effective transformation of data into information requires harmonization of stakeholder goals, implementation, enforcement of standards regarding quality and completeness, and availability of resources for maintenance and updates. Systems such as the Biomedical Data Translator leverage knowledge graphs (KGs), structured and machine learning readable knowledge representation, to encode knowledge extracted through inference. We focus here on the transformation of data from multiomics datasets and EHRs into compact knowledge, represented in a KG data structure. We demonstrate this data transformation in the context of the Translator ecosystem, including clinical trials, drug approvals, cancer, wellness, and EHR data. These transformations preserve individual privacy. We provide access to the five resulting KGs through the Translator framework. We show examples of biomedical research questions supported by our KGs, and discuss issues arising from extracting biomedical knowledge from multiomics data.