Project description:Amyloidoses are characterized by the pathological deposition of non-degradable misfolded protein fibrils. Precise identification of the fibril-forming protein is crucial for prognosis and correct therapeutic intervention. Here, we present a reproduceable method for amyloid typing using relative quantification to enhance the accuracy and reliability of proteomic amyloid typing. In this study, we analyzed 62 FFPE tissue samples (+4 replicates of one tissue sample using different set-ups), using liquid chromatography-tandem mass spectrometry and employed internal normalization of iBAQ values of amyloid-related proteins relative to serum amyloid-P component (APCS) for amyloidosis typing. This method demonstrated robust performance across multiple LC-MS/MS platforms, as well as for samples with low amounts of amyloid, and achieved complete concordance with IHC typed amyloidosis cases. More importantly, it resolved several unclear amyloid cases with inconclusive staining results. Finally, we established machine learning approach (XGBoost) achieving 94% accuracy by using ~160 amyloid-related proteins as input variables.
Project description:Purpose: This study uses a high-throughput glycan microarray to develop a novel method to assign ABO blood type. The method will then be applied to samples from patients treated with PROSTVAC to determine if blood type correlates with survival Results: Many blood group A and B antigens correlate with blood type. Blood typing is best achieved using a combination of 10 signals Conclusion: ABO blood type can be determined with greater than 97% accuracy using only 4 microliters of serum.
Project description:Chagas disease is one of the most important neglected diseases with an estimated number of 12 million infected individuals, the majority living in Central and South America. The Trypanosoma cruzi (T.cruzi) protozoan parasite is the etiological agent of Chagas disease. T.cruzi is highly genetically diverse and a new nomenclature assigned each strain to seven genetic groups (TcI-TcVI and Tcbat), named Discrete Typing Units (DTUs), based on their biochemical, immunological and phenotypical characteristics. T.cruzi DTUs have been correlated to diverse clinical outcomes highlighting the importance of molecular epidemiological screens. Despite the development of T.cruzi typing methods based on genetic signatures, each method presenting its own advantages and challenges. The work presented here shows the application of mass spectrometry for Trypanosoma cruzi Strain Typing Assay using MS2 peptide spectral libraries (Tc-STAMS2). The novelty of the method is based on the use of peptide fragmentation spectra as strain-specific fingerprints to classify and identify DTUs. Initially, a spectra library is generated from characterized T.cruzi strains. The library is subsequently inspected using MS/MS spectra from unknown strains and confidently assigned to a specific strain in an automated and computationally-driven approach. The Tc-STAMS2 method was challenged to test several variables such as sample type and preparation, instrument setup and identification platform. Tc-STAMS2 provided high confidence and robustness in T.cruzi strain typing. The Tc-STAMS2 method represents a proof-of-concept of a complementary strategy to the current DNA-based T. cruzi genotyping methods. Moreover, the method allows the identification of strain-specific features that could be related to the biology of T.cruzi strains and their clinical outcomes.
Project description:Comparison of High Resolution Viruelnce Allelic Profiling (HReVAP) typing with Multilocus Sequence Typing and Whole Genome SNPs analysis for typing VTEC strains