A machine learning-based diagnosis modeling of IgG4 Hashimoto's thyroiditis.

PURPOSE: This study aims to develop a non-invasive diagnosis model using machine learning (ML) for identifying high-risk IgG4 Hashimoto's thyroiditis (HT) patients. METHODS: A retrospective cohort of 93 HT patients and a prospective cohort of 179 HT patients were collected. According to the immunohistochemical and pathological results, the patients were divided into IgG4 HT group and non-IgG4 HT group. Serum TgAb IgG4 and TPOAb IgG4 were detected by ELISAs. A logistic regression model, support vector machine (SVM) and random forest (RF) were used to establish a clinical diagnosis model for IgG4 HT. RESULTS: Among these 272 patients, 40 (14.7%) were diagnosed with IgG4 HT. Patients with IgG4 HT were younger than those with non-IgG4 HT (P < 0.05). Serum levels of TgAb IgG4 and TPOAb IgG4 in IgG4 HT group were significantly higher than those in non-IgG4 HT group (P < 0.05). There were no significant differences in gender, disease duration, goiter, preoperative thyroid function status, preoperative TgAb or TPOAb levels, and thyroid ultrasound characteristics between the two groups (all P > 0.05). The accuracy, sensitivity, and specificity were 57%, 78%, and 79% for logistic regression model of IgG4 HT, 80 ± 7%, 84.7% ± 2.6%, and 75.4% ± 9.6% for the RF model and 78 ± 5%, 89.8% ± 5.7%, and 64.7% ± 5.7% for the SVM model. The RF model works better than SVM. The area under the ROC curve of RF ranged 0.87 to 0.92. CONCLUSION: A clinical diagnosis model for IgG4 HT established by RF model might help the early recognition of the high-risk patients of IgG4 HT.

Crystal Orientation in Pt-Based Alloys Induced by W(CO)6: Driving Oxygen Electroreduction Catalysis.

Integrating crystal orientation as well as structural and compositional advantages into one catalyst might be a promising strategy for high-performance Pt-based catalysts for proton-exchange membrane fuel cells. Herein, by introducing W(CO)6 as a structure-oriented template, Pt-based alloys with a well-defined crystal orientation along the (111) facet were obtained. The oxygen reduction reaction mass and specific activities of the crystal-facet-tuned alloys reach a new level. Moreover, the outstanding durability stems from the combination of their exposed crystal facets and incorporated W. The density functional theory calculation results reveal that the formation of the preferred (111) alloys can be attributed to the lower free energy of (111) facets and the weaker adsorption of CO released by W(CO)6. This proposed synthesis strategy of using transition-metal carbonyl compounds as additives to synthesize alloys with strong crystal orientation may open a door to the design of various alloy catalysts with ultrahigh activity.

Machine learning to predict clinical remission in depressed patients after acute phase selective serotonin reuptake inhibitor treatment.

OBJECTIVE: Selective serotonin reuptake inhibitors (SSRIs) are suggested as the first-line treatment for patients with major depressive disorder (MDD), but the remission rate is unsatisfactory. We aimed to establish machine learning models and explore variables available at baseline to predict the 8-week outcome among patients taking SSRIs. METHODS: Data from 400 patients were used to build machine learnings. The last observation carried forward approach was used to determine the remitter/non-remitter status of the patients at week 8. Using least absolute shrinkage and selection operator (LASSO) to select features, we built 4 different machine learning algorithms including gradient boosting decision tree, support vector machine (SVM), random forests, and logistic regression with five-fold cross-validation. Then, we adopted Shapley additive explanations (SHAP) values to interpret the model output. RESULTS: The remission rate is 67.8%. We obtained 78 features from the baseline characteristics, including 25 sociodemographic characteristics, 31 clinical features, 15 psychological traits and 7 neurocognitive functions, and 13 of these features were selected to establish SVM. The accuracy of the SVM prediction is 74.49%, reaching an average area under the curve of 0.734±0.043. The sensitivity is 0.899±0.038 with a positive predictive value of 0.776±0.028. The specificity is 0.422±0.091 with a negative predictive value of 0.674±0.086. According to the SHAP values, neurocognitive functions and anxiety and hypochondriasis symptoms were important predictors. CONCLUSION: Our study supports the utilization of machine learning approaches with inexpensive and highly accessible variables to accurately predict the 8-week treatment outcome of SSRIs in patients with MDD.

Glycolipid GD3 and GD3 synthase are key drivers for glioblastoma stem cells and tumorigenicity.

The cancer stem cells (CSCs) of glioblastoma multiforme (GBM), a grade IV astrocytoma, have been enriched by the expressed marker CD133. However, recent studies have shown that CD133(-) cells also possess tumor-initiating potential. By analysis of gangliosides on various cells, we show that ganglioside D3 (GD3) is overexpressed on eight neurospheres and tumor cells; in combination with CD133, the sorted cells exhibit a higher expression of stemness genes and self-renewal potential; and as few as six cells will form neurospheres and 20-30 cells will grow tumor in mice. Furthermore, GD3 synthase (GD3S) is increased in neurospheres and human GBM tissues, but not in normal brain tissues, and suppression of GD3S results in decreased GBM stem cell (GSC)-associated properties. In addition, a GD3 antibody is shown to induce complement-dependent cytotoxicity against cells expressing GD3 and inhibition of GBM tumor growth in vivo. Our results demonstrate that GD3 and GD3S are highly expressed in GSCs, play a key role in glioblastoma tumorigenicity, and are potential therapeutic targets against GBM.

Stage-specific embryonic antigen-4 as a potential therapeutic target in glioblastoma multiforme and other cancers.

Glioblastoma multiforme (GBM), the grade IV astrocytoma, is the most common and aggressive brain tumor in adults. Despite advances in medical management, the survival rate of GBM patients remains poor, suggesting that identification of GBM-specific targets for therapeutic development is urgently needed. Analysis of several glycan antigens on GBM cell lines revealed that eight of 11 GBM cell lines are positive for stage-specific embryonic antigen-4 (SSEA-4), and immunohistochemical staining confirmed that 38/55 (69%) of human GBM specimens, but not normal brain tissue, were SSEA-4(+) and correlated with high-grade astrocytoma. In addition, an SSEA-4-specific mAb was found to induce complement-dependent cytotoxicity against SSEA-4(hi) GBM cell lines in vitro and suppressed GBM tumor growth in mice. Because SSEA-4 is expressed on GBM and many other types of cancers, but not on normal cells, it could be a target for development of therapeutic antibodies and vaccines.

Redox regulation of the protein tyrosine phosphatase PTP1B in cancer cells.

The oxidation and inactivation of protein tyrosine phosphatases is one mechanism by which reactive oxygen species influence tyrosine phosphorylation-dependent signaling events and exert their biological functions. In the present study, we determined the redox status of endogenous protein tyrosine phosphatases in HepG2 and A431 human cancer cells, in which reactive oxygen species are produced constitutively. We used mass spectrometry to assess the state of oxidation of the catalytic cysteine residue of endogenous PTP1B and show that this residue underwent both reversible and irreversible oxidation to high stoichiometry in response to intrinsic reactive oxygen species production. In addition, our data show that the oxidation of PTP1B is specific to the active site Cys, with the other Cys residues in the protein remaining in a reduced state. Treatment of these cells with diphenyleniodonium, an inhibitor of NADPH oxidases, decreased reactive oxygen species levels. This resulted in inhibition of protein tyrosine phosphatase oxidation, concomitant with decreased tyrosine phosphorylation of cellular proteins and inhibition of anchorage-independent cell growth. Therefore, our data also suggest that the high level of intrinsic reactive oxygen species may contribute to the transformed phenotype of HepG2 and A431 cells via constitutive inactivation of cellular protein tyrosine phosphatases.