Project description:Background & Aims: Cholangiocarcinomas (CCAs), heterogeneous biliary tumors with dismal prognosis, lack accurate early-diagnostic methods, especially important for individuals at high-risk (i.e., primary sclerosing cholangitis (PSC)). Here, we searched for protein biomarkers in serum extracellular vesicles (EVs). Methods: EVs from patients with isolated PSC (n=45), concomitant PSC-CCA (n=42), PSC who developed CCA during follow-up (PSC to CCA; n=25), CCAs from non-PSC etiology (n=56), hepatocellular carcinoma (n=34) and healthy individuals (n=55) were characterized by mass-spectrometry. Diagnostic biomarkers of PSC-CCA, non-PSC CCA or CCAs regardless etiology (pan-CCAs) were defined, and their expression was evaluated in human organs/tissues and within CCA tumors at single-cell level. Prognostic EV-biomarkers for CCA were investigated. Results: High-throughput proteomics identified candidate diagnostic biomarkers for PSC-CCA, non-PSC CCA or pan-CCA, as well as and for differential diagnosis of intrahepatic CCA and HCC, that were cross-validated by ELISA using total serum. Machine learning logit modelling disclosed CRP/FRIL/Fibrinogen algorithm with diagnostic value for early-stage PSC-CCA vs isolated PSC (AUC=0.944; OR=82.0), overpowering CA19-9 (AUC=0.735; OR=9.3). An algorithm combining CRP/VWF/PIGR/ /Fibrinogen allowed the diagnosis of early-stage non-PSC CCAs compared to healthy individuals (AUC=0.999; OR=1115). Noteworthy, levels of Fibrinogen/CRP/PIGR/FRIL showed predictive capacity for CCA development in patients with PSC before clinical evidences of malignancy. Multi-organ transcriptomic analysis revealed that serum EV-biomarkers were mostly expressed in hepatobiliary tissues, and scRNA-seq and immunofluorescence analysis of CCA tumors showed their presence mainly in malignant cholangiocytes. Multivariable analysis unveiled EVprognostic biomarkers independent to clinical features, with COMP/GNAI2/CFAI and ACTN1/MYCT1/PF4V associated negatively or positively to patients’ survival, respectively. Conclusions: Serum EVs contain protein biomarkers for the prediction, early diagnosis and prognosis estimation of CCA, representing a novel tumor cell-derived liquid biopsy for personalized medicine.

Project description:In gas chromatography-mass spectrometry-based untargeted metabolomics, metabolites are identified by comparing mass spectra and chromatographic retention time with reference databases or standard materials. In that sense, machine learning has been used to predict the retention time of metabolites lacking reference data. However, the retention time prediction of trimethylsilyl derivatives of metabolites, typically analyzed in untargeted metabolomics using gas chromatography, has been poorly explored. Here, we provide a rationalized framework for machine learning-based retention time prediction of trimethylsilyl derivatives of metabolites in gas chromatography. We compared different machine learning paradigms, in addition to exploring the influence of the computational molecular structure representation to train the prediction models: fingerprint class and fingerprint calculation software. Our study challenged predicted retention time when using chemical ionization and electron impact ionization sources in simulated and real cases, demonstrating a good correct identity ranking capability by machine learning, despite observing a limited false identity filtering power in cases where a spectrum or a monoisotopic mass match to multiple candidates. Specifically, machine learning prediction yielded median absolute and relative retention index (relative retention time) errors of 37.1 retention index units and 2%, respectively. In addition, fingerprint class and fingerprint calculation software, as well as the molecular structural similarity between the training and test or real case sets, showed to be critical modulators of the prediction performance. Finally, we leveraged the structural similarity between the training and test or real case set to determine the probability that the prediction error is below a specific threshold. Overall, our study demonstrates that predicted retention time can provide insights into the true structure of unknown metabolites by ranking from the most to the least plausible molecular identity, and sets the guidelines to assess the confidence in metabolite identification using predicted retention time data.

			Action	DRS
	B10a.raw	Raw
	B10a.raw.-1.mgf	Mgf
	B10b.raw	Raw
	B10b.raw.-1.mgf	Mgf
	B10c.raw	Raw

Dataset Information

Computational analysis of cholangiocarcinoma phosphoproteomes identifies patient-specific drug targets

Dataset's files

Publications

Computational Analysis of Cholangiocarcinoma Phosphoproteomes Identifies Patient-Specific Drug Targets.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets