Project description:Artificial intelligence (AI) applications in biomedical settings face challenges such as data privacy and regulatory compliance. Federated Deep Learning (FDL) effectively addresses these issues. We developed ProCanFDL, where local models were trained on simulated sites using proteomic data drawn from a pan-cancer cohort (n = 1,260) and 29 other cohorts (n = 6,265), representing 4,956 patients and 19,930 mass spectrometry (MS) runs, all held behind private firewalls. Local parameter updates were aggregated to build the global model, achieving a 43% performance gain over local models on the hold-out test set (n = 625) in 14 cancer subtyping tasks. Additionally, ProCanFDL preserved data privacy while matching centralized model performance. External validation assessed generalization by retraining the global model with data from two external cohorts (n = 55) and eight (n = 832) using a different MS technology. ProCanFDL presents a solution for internationally collaborative machine learning initiatives using proteomic data while maintaining data privacy.
Project description:we analyze expression of miRNAs in a cohort of male and female patients with familial breast cancer (BRCA1/2-related and BRCAX) and in a subset of sporadic breast cancer
Project description:Breast cancer was one of the first cancer types where molecular subtyping led to explanation of interpersonal heterogeneity and resulted in improvement of treatment regimen. Several multigene classifiers have been developed and in particular those defining molecular signatures of early breast cancers possess significant prognostic information. Hence since 2014, molecular subtyping of primary breast cancers was implemented as a part of routine diagnostics with direct impact of therapy assignment. In this study, we evaluate direct and potential benefits of molecular subtyping in low-risk breast cancers as well as present the advantages of a robust molecular signature in regard to patient work-up among high-risk breast cancers.
Project description:RNA-seq was performed on breast cancer cell lines and primary tumors RNA-seq was performed on 28 breast cancer cell lines, 42 Triple Negative Breast Cancer (TNBC) primary tumors, and 42 Estrogen Receptor Positive (ER+) and HER2 Negative Breast Cancer primary tumors, 30 uninovlved breast tissue samples that were adjacent to ER+ primary tumors, 5 breast tissue samples from reduction mammoplasty procedures performed on patients with no known cancer, and 21 uninvolved breast tissue samples that were adjacent to TNBC primary tumors.
Project description:This SuperSeries is composed of the following subset Series:; GSE6604: Expression data from Normal Prostate Tissue free of any pathological alteration; GSE6605: Expression data from Metastatic Prostate Tumor; GSE6606: Expression data from Primary Prostate Tumor; GSE6608: Expression data from Normal Prostate Tissue Adjacent to Tumor Experiment Overall Design: Refer to individual Series
Project description:Ten arrays were performed on the RNA extracts from 10 patients' samples, each of them contained the paired samples tumor tissue/ normal adjacent tissue.
Project description:Despite improvements in capabilities of proteomics technologies, the introduction of new plasma-based protein biomarkers for clinical use remains low. One reason is the cumbersome requirement to test thousands of protein candidates in follow-up quantitative verification studies. We sought to evaluate internal standard triggered parallel reaction monitoring (IS-PRM) in the context of biomarker verification by developing a method to quantify 5,176 peptides (1,314 proteins) as candidate biomarkers for early detection of breast cancer. Method performance was characterized in a response curve showing large linear range (4 orders of magnitude) and good repeatability (median CV 7.7%). The method was applied to pools of cancer and control human plasma, detecting 893 proteins and qualifying 164 candidates to advance for further evaluation. The method shows good quantitative performance, greatly expanding the capabilities for quantification of large numbers of proteins, and is well suited for large scale relative quantification of protein sets.
Project description:The miRNA expression profiles in one pair of hTERT-positive gastric cancer tissue and an hTERT-negative para-cancerous tissue. The para-cancerous tissue is at least 5cm away from the cancer tisse. The expression of hTERT of identified by immunohistochemistry before RNA extraction for miRNA assay. One pair of gastric cancer tissue and para-cancerous tissue(Control). Four replicates per array.