With the development of mass spectrometry technologies and bioinformatics analysis algorithms, disease research-driven human proteome project (HPP) is advancing rapidly. Protein biomarkers play critical roles in clinical applications and the biomarker discovery strategies and methods have become one of research hotspots. Feature selection and machine learning methods have good effects on solving the "dimensionality" and "sparsity" problems of proteomics data, which have been widely used in the discovery of protein biomarkers. Here, we systematically review the strategy of protein biomarker discovery and the frequently-used machine learning methods. Also, the review illustrates the prospects and limitations of deep learning in this field. It is aimed at providing a valuable reference for corresponding researchers.
Algorithms ; Biomarkers ; Humans ; Machine Learning ; Mass Spectrometry ; Proteomics

Objective:To investigate the influence of frailty on post-treatment outcomes in elderly heart failure patients with reduced ejection fraction treated with Sacubitril/Valsartan.Methods:The 231 heart failure patients aged 60 years or over with reduced ejection fraction were enrolled from October 2017 to October 2018 in Department of Geriatric Medicine, Henan Provincial People's Hospital.Patients were divided into the frailty group(n=116)and the control group(n=115). Frailty diagnosis was made by five indexes suggested by LP Fried.Both groups were treated with sacubitril/valsartan(49/51 mg)for 1 year.The left ventricular ejection(LVEF), estimated glomerular filtration rate(eGFR), N-terminal pro B-type natriuretic peptide(NT-proBNP)and other clinical and laboratory indexes were detected before and after treatment and compared between the frailty group and the control group.Results:16 subjects in the frailty group and 11 subjects in the control group dropped out of the study.The frailty group versus the control group showed a higher mortality rate of cardiovascular causes(13.0% or 13/100 vs.6.7% or 7/104, χ2=6.437, P=0.027), a higher first re-hospitalization rate(18.0% or 18/100 vs.11.5% or 12/104, χ2=4.458, P=0.043)and a higher all-cause mortality(16.0% or 16/100 vs.8.6% or 9/104, χ2=3.875, P=0.039). In the frailty group, levels of serum NT-proBNP and creatinine were higher and eGFR was lower after treatment than before treatment[(2 253±144) ng/L vs.(2 094±136) ng/L, (137±24) μmol/L vs.(125±23) μmol/L, (49.2±5.9) ml·min -1·1.73 m -2vs.(56.7±6.3) ml·min -1·1.73 m -2, t=3.674, 2.893 and 2.068, P=0.017, 0.026 and 0.029]. In the control group, serum NT-proBNP levels were lower after treatment than before treatment[(1 828±123) ng/L vs.(1 945±128) ng/L, t=1.896, P=0.043], while serum creatinine levels[(120±22) μmol/L vs.(117±19) μmol/L, t=2.099, P=0.650]and eGFR[(59.8±6.5) ml·min -1·1.73 m -2vs.(61.6±6.8) ml·min -1·1.73 m -2, t=2.444, P=0.173]had no significant difference between post-treatment and pre-treatment. Conclusions:Frailty has adverse affects on the mortality, re-hospitalization rate and renal function in elderly heart failure patients with reduced ejection fraction treated with Sacubitril/Valsartan.

Objective:To investigate the association of time in range with metabolic associated fatty liver disease(MAFLD) and advanced liver fibrosis in patients with type 2 diabetes.Methods:This study was a retrospective study. A total of 494 type 2 diabetic patients were recruited in the Department of Endocrinololgy of Henan Provincial People′s Hospital from November 2019 to April 2022. Time in range(TIR) was calculated with continuous glucose monitoring data. Abdominal ultrasound scan was used to diagnose fatty liver. Liver stiffness measurement(LSM) by transient elastography was used to evaluate liver fibrosis. Pearson and multivariate linear regression analysis was used to evaluate the association between TIR and LSM. Multivariate logistic regression analysis was used to analyze the association of TIR with risk of MAFLD and advanced liver fibrosis.Results:Pearson correlation analysis showed that LSM was negatively correlated with TIR( r=-0.86, P<0.001) and was positively correlated with homeostasis model assessment for insulin resistance(HOMA-IR; r=0.48, P<0.001). After adjusting for confounding factors, multivariate linear regression analysis showed that TIR significantly negatively predicted LSM( β=-0.75, P<0.001), and HOMA-IR significantly positively predicted LSM( β=0.21, P=0.025). After adjusting for confounding factors, logistic regression analysis showed that compared with TIR Q4 patients, TIR Q1 patients had an increased risk of MAFLD( OR=1.96, 95% CI 1.07-3.62, P=0.027), advanced liver fibrosis( OR=3.82, 95% CI 1.17-12.50, P=0.027), and HOMA-IR was an independent risk factor for MAFLD( OR=1.22, 95% CI 1.04-1.43, P=0.005) and advanced liver fibrosis( OR=1.26, 95% CI 1.03-1.54, P=0.025). Conclusions:TIR and insulin resistance are independent risk factors for MAFLD and advanced liver fibrosis in patients with type 2 diabetes. TIR has a significant predictive value for MAFLD and advanced liver fibrosis.