Software Options for the Analysis of MS-Proteomic Data

Proteomics strategies for urothelial bladder cancer diagnosis, prognosis and treatment: Trends for tumor biomarker sources

Urothelial bladder cancer (UBC) is a heterogeneous multifactorial malignancy with a high recurrence rate. Current procedures for UBC diagnosis suffering from the lack of clinical sensitivity and specificity screening tests. Therefore, biomarkers have promising values to predict pathological conditions and can be considered as effective targets for early diagnosis, prognosis and antitumor immunotherapy. Recently, researchers have been interested for tumor proteins as biomarkers for different diseases. At present, proteomics methods have rapidly progressive that has potential identified biomarkers of UBC. Specifically, there has been several studies on the potential application of proteomics for the identification, quantification, and profiling of proteins for UBC in different sources. Based on these studies, using the panel of biomarkers as proteomic patterns may achieve higher sensitivity and specificity than single proteins in the diagnosis of UBC. In the present review, we evaluate recent literature related to the UBC proteome focusing especially on new proteomics techniques. Moreover, we classify UBC tumor biomarkers as diagnostic, prognostic, and therapeutic targets based on their sources (urine, serum/plasm, cell line, and tumor tissue) and we also discuss the advantages and limitations of each source. In this manner, this review article provides a critical assessment presentation of the advances in proteomics for all aspects of UBC diagnosis, prognosis, and treatment based on sources.

Variability analysis of LC-MS experimental factors and their impact on machine learning

BACKGROUND

Machine learning (ML) technologies, especially deep learning (DL), have gained increasing attention in predictive mass spectrometry (MS) for enhancing the data-processing pipeline from raw data analysis to end-user predictions and rescoring. ML models need large-scale datasets for training and repurposing, which can be obtained from a range of public data repositories. However, applying ML to public MS datasets on larger scales is challenging, as they vary widely in terms of data acquisition methods, biological systems, and experimental designs.

RESULTS

We aim to facilitate ML efforts in MS data by conducting a systematic analysis of the potential sources of variability in public MS repositories. We also examine how these factors affect ML performance and perform a comprehensive transfer learning to evaluate the benefits of current best practice methods in the field for transfer learning.

CONCLUSIONS

Our findings show significantly higher levels of homogeneity within a project than between projects, which indicates that it is important to construct datasets most closely resembling future test cases, as transferability is severely limited for unseen datasets. We also found that transfer learning, although it did increase model performance, did not increase model performance compared to a non-pretrained model.