DNAmBERT: A Transformer-Based Model for Robust, Non-Invasive Cancer Diagnosis Using DNA Sequence and Methylation Data
Ontology highlight
ABSTRACT:
ORGANISM(S): Homo sapiens
PROVIDER: GSE301266 | GEO | 2026/06/26
REPOSITORIES: GEO
Similar Datasets
Project description:Lung cancer is the most common cause of death from cancer worldwide, largely due to frequent late diagnosis. Thus, there is the urgent need to develop new approaches to improve theearly detection of lung cancer, which would greatly affect patient survival. The quantitative protein expression profiles of microvesicles isolated from sera from 46 lung cancer patients and 41 high-risk non-cancer subjects were obtained using a mass spectrometry method based on library matching. We identified 33 differentially expressed proteins that allow discriminating the two groups. Based on serum protein expression profiles, we built a machine learning algorithm , which highlighted a decrease in the levels of Arysulfatase A (ARSA) as the most discriminating factor found in tumors. Our study identified a signature able to discriminate with high specificity and selectivity early lung cancer patients from high-risk healthy subjects. Among the differentially expressed proteins, ARSA emerged as the most promising biomarker for lung cancer early diagnosis.
2023-02-13 | PXD034221 | Pride
Project description:Background: Lung cancer (LC) is the leading cause of cancer deaths worldwide with more than 1.7 million deaths each year. Early detection of LC may be crucial to achieve efficient treatment and to increase survival. However, there are limitations to current methods for early detection of LC as they are too invasive and represent potential health hazards, necessitating improved tools to detect LC at an early stage. Blood-based gene expression profiling is an alternative, or additional tool for diagnostic purposes as blood samples are easily available, essentially non-invasive, and can be collected at a low cost. However, as the blood collection method can affect measured gene expression, the aim of this study was to identify mRNAs that are robust biomarkers for LC in blood. How: We sampled blood from 123 LC patients at the St. Olavs university hospital, and 180 controls from two different biobanks sampled on two different blood tubes (PAXgene and Tempus) and used Illumina (HT-12 v4) microarrays to measure whole blood gene expression. We did three analyses: (i) finding relevant gene expression differences between cases and controls, (ii) finding a robust set of genes to be used as a panel to identify presence of LC in blood, and (iii) identifying differences between patients with different pathological traits such as stage and histology. We collected phenotype data from questionnaires and hospital medical records, and evaluated the potential effects of CRP, tumour size, gender, and smoking habits. Results: By comparing cases and controls sampled on the same RNA sampling system (training set), we identified 355 significant genes (Bonferroni adjusted p < 0.05) with biological relevance. When evaluating these on our test set comprising of the same cases as used in the training set but analysed against controls from a different biobank sampled on a different RNA sampling system (test set), we found 50 genes that were robust as they were unaffected by either technical issues in sampling systems, biobank differences, gender, or pathological traits such as LC subtype or stadium. These findings were confirmed in three validation sets and the robust list’s diagnostic potential was validated on RNA sequencing data from a new cohort. By analysing the cases separately, we found seven genes distinguishing squamous cell carcinoma (SCC) from other LC subtypes, and six genes distinguishing early from late-stage LC patients. Pathway analyses and a literature survey indicated that the identified genes show biological relevance for cancer development and lung related diseases.
2024-12-10 | GSE252168 | GEO
Project description:Lung cancer remains the most common cause of cancer deaths worldwide, yet there is currently a lack of diagnostic noninvasive biomarkers that could guide treatment decisions. Small molecules (<1500 Da) were measured in urine collected from 469 lung cancer patients and 536 population controls using unbiased liquid chromatography-mass spectrometry. Clinical putative diagnostic and prognostic biomarkers were validated by quantitation and normalized to creatinine levels at two different time points and further validated in an independent sample set, which comprises 80 cases and 78 population controls, with similar demographic and clinical characteristics when compared to the training set. Creatine riboside (IUPAC name: 2-{2-[(2R,3R,4S,5R)-3,4-dihydroxy-5-(hydroxymethyl)-oxolan-2-yl]-1-methylcarbamimidamido}acetic acid), a novel molecule identified in this study, and N-acetylneuraminic acid (NANA), were each significantly (P <0.00001) elevated in non–small cell lung cancer (NSCLC) and associated with worse prognosis (hazard ratio (HR) =1.81 [P =0.0002], and 1.54 [P =0.025], respectively). Creatine riboside was the strongest classifier of lung cancer status in all and stage I–II cases, important for early detection, and also associated with worse prognosis in stage I–II lung cancer (HR =1.71, P =0.048). All measurements were highly reproducible with intraclass correlation coefficients ranging from 0.82 – 0.99. Both metabolites were significantly (P <0.03) enriched in tumor tissue compared to adjacent non-tumor tissue (N =48), thus revealing their direct association with tumor metabolism. Creatine riboside and NANA may be robust urinary clinical metabolomic markers that are elevated in tumor tissue and associated with early lung cancer diagnosis and worse prognosis.
2014-05-06 | MTBLS28 | MetaboLights
Project description:Lung cancer is the leading cause of cancer mortality and early detection is the key to improve survival. However, there are no reliable blood-based tests currently available for early-stage lung cancer diagnosis. Here, we performed single-cell RNA sequencing of early-stage lung cancer and found lipid metabolism was broadly dysregulated in different cell types and glycerophospholipid metabolism is the most significantly altered lipid metabolism-related pathway. Untargeted lipidomics were detected in an exploratory cohort of 311 participants. Through support vector machine algorithm-based and mass spectrum-based feature selection, we have identified nine lipids as the most important detection features and developed a LC-MS-based targeted assay utilizing multiple reaction monitoring. This target assay achieved 100.00% specificity on an independent validation cohort. In a hospital-based lung cancer screening cohort of 1036 participants examined by low dose CT and a prospective clinical cohort containing 109 participants, this assay reached over 90.00% sensitivity and 92.00% specificity. Accordingly, matrix-assisted laser desorption/ionization-mass spectrometry imaging assay confirmed the selected lipids were differentially expressed in early-stage lung cancer tissues in situ. Thus, this method, designated as Lung Cancer Artificial Intelligence Detector (LCAID), may be used for early detection of lung cancer or large-scale screening of high-risk populations in cancer prevention.
2021-11-05 | MSV000088324 | GNPS
Project description:Plasma samples from 100 early stage (I to IIIA) non–small-cell lung cancer (NSCLC) patients and 100 non-cancer controls were screened for 754 circulating microRNAs via qRT-PCR, using TaqMan MicroRNA Arrays. Our objective was to identify a panel of circulating microRNAs in plasma that will contribute to early detection of lung cancer.
2015-12-31 | GSE64591 | GEO
Project description:Lung cancer is the second most commonly diagnosed cancer and the leading cause of cancer death worldwide, of which approximately 85% are non-small cell lung cancer (NSCLC). The overall survival (OS) of patients with advanced NSCLC was significantly prolonged with immune checkpoint inhibitors (ICIs) targeting the programmed cell death-1 (PD-1) and programmed death-ligand 1 (PD-L1) axis. For early-stage lung cancer, the 5-year survival rate for patients ranges from 80% in stage IA to 41% in stage IIIA, and many cases relapse after surgical resection. Currently, multiple clinical trials have manifested the encouraging efficacy of neoadjuvant immunotherapy in stage I-IIIA resectable NSCLC. However, the effect of immunotherapy in ultra early-stage NSCLC patients with micro-invasive or even pre-invasive lesions remains unclear. In this study, we aimed to evaluate the activity and safety of sintilimab on high-risk ground glass opacity lesions in multiple primary lung cancer patients.
2024-05-01 | GSE260770 | GEO
Project description:Gastric cancer (GC) remains one of the leading causes of cancer-related death worldwide, especially in East Asian countries. Despite advances in treatment, patients with locally advanced gastric cancer (LAGC) still have a poor prognosis due to the high incidence of metastasis and recurrence. Peritoneal metastasis is particularly occult and difficult to detect by conventional imaging techniques at an early stage. Existing imaging methods lack high sensitivity and are difficult to detect early metastatic lesions. Although diagnostic laparoscopy has high sensitivity, it is invasive and difficult to use routinely. Therefore, innovative diagnostic methods are urgently needed. Transcriptomics methods can provide a comprehensive understanding of the biological state of cancer, especially showing potential in the early detection of GC and peritoneal metastasis. Our objective is to provide a more sensitive diagnostic method in the latent stage of peritoneal metastasis by identifying the RNA expression pattern of cancer cells.
2025-04-23 | GSE289037 | GEO
Project description:Lung cancer is the leading cause of cancer mortality and early detection is the key to improve survival. However, there are no reliable blood-based tests currently available for early-stage lung cancer diagnosis. Here, we performed single-cell RNA sequencing of early-stage lung cancer and found lipid metabolism was broadly dysregulated in different cell types and glycerophospholipid metabolism is the most significantly altered lipid metabolism-related pathway. Untargeted lipidomics were detected in an exploratory cohort of 311 participants. Through support vector machine algorithm-based and mass spectrum-based feature selection, we have identified nine lipids as the most important detection features and developed a LC-MS-based targeted assay utilizing multiple reaction monitoring. This target assay achieved 100.00% specificity on an independent validation cohort. In a hospital-based lung cancer screening cohort of 1036 participants examined by low dose CT and a prospective clinical cohort containing 109 participants, this assay reached over 90.00% sensitivity and 92.00% specificity. Accordingly, matrix-assisted laser desorption/ionization-mass spectrometry imaging assay confirmed the selected lipids were differentially expressed in early-stage lung cancer tissues in situ. Thus, this method, designated as Lung Cancer Artificial Intelligence Detector (LCAID), may be used for early detection of lung cancer or large-scale screening of high-risk populations in cancer prevention.
2021-11-05 | MSV000088324 | MassIVE
Project description:We identified candidate protein biomarkers to distinguish lung adenocarcinomas from benign nodules. We employed shotgun proteomics using a multidimensional peptide separation approach coupled to tandem mass spectrometry (LC-MS/MS) to characterize proteomes of a collection of 34 benign lung nodules. The benign lung nodule inventories were compared to inventories from early stage lung adenocarcinomas, from 24 patients with non-adjuvant treatment following the resection, and to inventories from normal tissue collected from 10 patients as bronchial and alveolar epithelium to identify biomarkers.We developed PRM assays to quantify a subset of 43 candidate biomarker proteins via 170 tryptic peptides in a single LC-PRM-MS run using a validation cohort of 20 benign nodules, 20 adenocarcinoma, and 20 normal lung tissues.
2016-10-09 | MSV000080236 | MassIVE
Project description:Whole exome sequencing of tumors and paired adjacent uninvolved tissues from 222 early stage NSCLC patients, in order to identify genomic drivers present in early-stage non-small cell lung cancer and determine the overall tumor mutational burden in early-stage non-small cell lung cancer.
| EGAD00001005956 | EGA