Electrospray triple quadrupole mass spectrometry guides pathologists to suggest appropriate molecular testing in the identification of rare hemoglobin variants.

Background: The presumptive diagnosis of hemoglobinopathies relies on routine tests such as Complete Blood Count (CBC), peripheral blood smear, Liquid Chromatography (LC), and Capillary Electrophoresis (CE), along with clinical findings. Pathologists suggest molecular sequencing of HBA and HBB genes to correlate blood picture with clinical findings in order to identify unknown rare haemoglobin (Hb) variants or variants that coelute with Hb. This paper presents a low-resolution mass spectrometry (MS)-based method for presumptive identification of variants that eluted in zone 12 of CE, followed by molecular sequencing of the HBB gene for a definitive diagnosis of hemoglobinopathies. Methods: Eight patient samples with a variant peak in zone 12 of CE (Sebia) were analyzed using MS. The mass-to-charge ratio (m/z) observed was deconvoluted to determine the mass of Hb variants. The ß variants were subsequently confirmed through molecular sequencing. Results: Based on the intact mass of the variants, there were two samples of the α variant (α + 58 Da and α + 44 Da), and six samples of the ß variant. Out of these six ß variant samples, three were the ß + 58 Da variant, and three were the ß + 30 Da variant. By correlating the intact mass information with the CE pattern and considering the ethnicity of the patients, it was presumed that the α variants were HbJ Meerut (α + 58 Da, x-axis 102) and HbJ Paris-I (α + 44 Da, x-axis 80). Molecular analysis confirmed the identity of ß variants as Hb Rambam/HbJ Cambridge, HbJ Bangkok (+58 Da), and Hb Hofu (+30 Da). Conclusion: The mass information of Hb variants obtained using Electrospray triple quadrupole MS assists pathologists in recommending the appropriate molecular sequencing for identifying unknown variants.

A Qualitative Method to Detect Paraproteins from Serum Using Ultra Performance Liquid Chromatography Electrospray Triple Quadrupole Mass Spectrometry.

BACKGROUND: Mass spectrometry-based techniques are increasingly reported in the literature for identifying paraproteins due to their improved specificity and sensitivity. The present study demonstrates the capability of ultra performance liquid chromatography (UPLC) electrospray ionization triple quadrupole mass spectrometry for the qualitative analysis of paraproteins. METHODS: Paraproteins from patient serum (n = 40) were immunopurified using agarose beads coated with camelid antibodies that are specific for various subtypes of immunoglobulins (Igs; G, A, M, and light chains κ, λ). The extracted Igs are reduced to separate light chains from heavy chains in solution. The reduced sample was subjected to UPLC and mass measured using electrospray ionization-mass spectrometry. The mass spectral peaks at specific retention times were deconvoluted after clean-up to obtain the mass of light chains. The interpretation of liquid chromatography peaks and LC-MS data was validated by comparing them with immunofixation electrophoresis (IFE) results. RESULTS: The interpretation from the chromatographic pattern had a 92.5% (37/40) agreement when compared with mass information. The correlation of mass spectrometry data to IFE was 90% (36/40). The high mass of light chains (>25 kDa) was suggestive of glycosylation. Patient sera positive for IgGκ on IFE (n = 15) were analyzed for the interference of tAbs. The mass of Daratumumab observed in a sample was confirmed by the treating physician. A biclonal of same isotype (IgGκ) was identified. CONCLUSIONS: The feasibility of using liquid chromatography triple quadrupole mass spectrometry for the identification of the subtype of paraproteins has been demonstrated. The method's applicability to screen for interference from tAbs and identification of biclonals of the same isotype has been highlighted.

Model-agnostic explainable artificial intelligence tools for severity prediction and symptom analysis on Indian COVID-19 data.

Introduction: The COVID-19 pandemic had a global impact and created an unprecedented emergency in healthcare and other related frontline sectors. Various Artificial-Intelligence-based models were developed to effectively manage medical resources and identify patients at high risk. However, many of these AI models were limited in their practical high-risk applicability due to their "black-box" nature, i.e., lack of interpretability of the model. To tackle this problem, Explainable Artificial Intelligence (XAI) was introduced, aiming to explore the "black box" behavior of machine learning models and offer definitive and interpretable evidence. XAI provides interpretable analysis in a human-compliant way, thus boosting our confidence in the successful implementation of AI systems in the wild. Methods: In this regard, this study explores the use of model-agnostic XAI models, such as SHapley Additive exPlanations values (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME), for COVID-19 symptom analysis in Indian patients toward a COVID severity prediction task. Various machine learning models such as Decision Tree Classifier, XGBoost Classifier, and Neural Network Classifier are leveraged to develop Machine Learning models. Results and discussion: The proposed XAI tools are found to augment the high performance of AI systems with human interpretable evidence and reasoning, as shown through the interpretation of various explainability plots. Our comparative analysis illustrates the significance of XAI tools and their impact within a healthcare context. The study suggests that SHAP and LIME analysis are promising methods for incorporating explainability in model development and can lead to better and more trustworthy ML models in the future.