Journal: BMC Biology

Article Title: Machine learning combined with omics-based approaches reveals T-lymphocyte cellular fate imbalance in abdominal aortic aneurysm

doi: 10.1186/s12915-025-02400-x

Figure Lengend Snippet: Identification of TIRS regulatory mechanisms and key biomarkers. A LASSO-based feature selection, with the optimal lambda determined when the partial likelihood deviance reached the minimum value (left). SVM-RFE-based feature selection, with root mean square error (RMSE) reached the minimum value and R -squared reached the max value (mid). Venn diagram presented the intersection of key biomarkers obtained through both algorithms (right). B Aberrant expression profiles for key biomarkers in Abdominal Aortic Wall Dataset 1 (AAA n = 80 patients, control n = 10 healthy individuals; Student’s t -test). C ROC curve demonstrating the diagnostic efficacy of FOSB, JUNB, CST7, and TBC1D4 in Abdominal Aortic Wall Dataset 1. D Clinical impact plot illustrating the clinical utility of key biomarkers. The “Number high risk” curve closely aligns with the “Number high risk with the event” curve at each threshold probability, indicating exceptional predictive power. E Aberrant expression profiles for key biomarkers in Abdominal Aortic Wall Dataset 2 (AAA n = 9 patients, control n = 10 healthy individuals; Student’s t -test). F ROC curve validating the diagnostic efficacy of FOSB, JUNB, CST7, and TBC1D4 in Abdominal Aortic Wall Dataset 2. G Clinical impact plot demonstrating the clinical utility of key biomarkers. Again, the “Number high risk” curve is closely aligned with the “Number high risk with the event” curve at each threshold probability, highlighting the biomarkers’ strong predictive power. H Aberrant expression profiles for key biomarkers in Perivascular Adipose Tissue Dataset 3 (dilated n = 30, non-dilated n = 30; Student’s t -test). I ROC curve verifying the diagnostic efficacy of FOSB, JUNB, CST7, and TBC1D4 in Perivascular Adipose Tissue Dataset 3. J Impact plots reiterated superior predictive performance probability, indicating outstanding predictive capability

Article Snippet: The primary antibodies against FOSB (1:500, catalog No. ab184938, Abcam) and JUNB (1:50, catalog No. 10486–1-AP, Proteintech) were incubated overnight at 4 °C, followed by incubation with horseradish peroxidase conjugated secondary antibodies.

Techniques: Selection, Expressing, Control, Diagnostic Assay

Journal: BMC Biology

Article Title: Machine learning combined with omics-based approaches reveals T-lymphocyte cellular fate imbalance in abdominal aortic aneurysm

doi: 10.1186/s12915-025-02400-x

Figure Lengend Snippet: Verification of key biomarkers. A Abdominal aortic wall and peripheral blood samples obtained from AAA patients. B Aberrant expression profiles for key biomarkers in the abdominal aortic wall (Inhouse Dataset 1; AAA n = 5 patients, control n = 4 healthy individuals). C ROC curve validating the diagnostic efficacy of FOSB, JUNB, CST7, and TBC1D4 in the abdominal aortic wall (Inhouse Dataset 1). D Clinical impact plot demonstrating the clinical utility of key biomarkers. The “Number high risk” curve closely aligns with the “Number high risk with the event” curve at each threshold probability, indicating exceptional predictive power. E Aberrant expression profiles for key biomarkers in peripheral blood (Inhouse Dataset 2; AAA n = 24 patients, control n = 15 healthy individuals). F ROC curve validating the diagnostic efficacy of FOSB, JUNB, CST7, and TBC1D4 in peripheral blood (Inhouse Dataset 2). G Clinical impact plot illustrating the clinical utility of key biomarkers. Again, the “Number high risk” curve remains closely aligned with the “Number high risk with the event” curve at each threshold probability, highlighting the biomarkers’ strong predictive capability. H Mice were infused with saline or Ang II (1000 ng/kg/min) + BAPN. Gross abdominal aorta images were shown. Scale bar is 1 cm. I Representative images of immunohistochemical stains for elastin fiber (Van Gieson) and representative photomicrographs of hematoxylin and eosin (H&E) staining. Scale bar is 200 μm. J – L Representative immunohistochemical staining of FOSB and JUNB in aortic cross sections. Scale bar is 50 μm. Data are expressed as mean ± SEM (control n = 3 mice, AAA n = 5 or 6 mice). Student’s t -test was utilized to compare continuous variables between the two groups

Article Snippet: The primary antibodies against FOSB (1:500, catalog No. ab184938, Abcam) and JUNB (1:50, catalog No. 10486–1-AP, Proteintech) were incubated overnight at 4 °C, followed by incubation with horseradish peroxidase conjugated secondary antibodies.

Techniques: Expressing, Control, Diagnostic Assay, Saline, Immunohistochemical staining, Staining