| Background:Inflammatory bowel disease (IBD), including Crohn’s disease (CD) and Ulerative Colitis (UC), is a chronic, inflammatory intestinal disease without definite etiology and pathogenesis. Due to lacking diagnostic golden standard, missed diagnosis or misdiagnosis exist in clinical practice. Moreover, the differential diagnoses between subtype of IBD (CD and UC), between IBD and Intestinal Tuberculosis (ITB), are still challenges. Serum proteomic fingerprint, which aims to explore disease-related proteins, study protein expression or change after modification, look for diagnostic markers and establish diagnostic models, is a new type of proteomics technology developed in recent years. This study, using weak cationic magnetic beads combined with Matrix-assisted Laser Desorption/Ionization Time Of Flight Mass Spectrometry (MALDI-TOF-MS) technique to analyse serum proteome of IBD (including CD and UC), ITB patients and healthy controls, based on the genetic algorithm combined with support vector machine (SVM) model, selecting the best diagnostic models and validated with Leave One Out (LOO) method, expects to provide new ideas about the early diagnoses of IBD and the differential diagnoses between CD and UC, CD and ITB at serum protein level.Objective:To explore the diagnostic models of IBD, the differential diagnostic models between CD and UC, between CD and ITB and their clinical implicational value by analyzing serum proteome profiles.Methods:Part one:serum proteome profiles from 108 IBD patients including 59 CD patients and 49 UC patients and 48 healthy controls were analyzed using weak cationic magnetic beads combined with MALDI-TOF-MS technique to detect the differentially expressed proteins of the serum samples. Three groups were compared:group of CD patients and healthy controls, group of UC patients and healthy controls, group of CD patients and UC patients. Wilcoxon rank sum test was used to screen the most differentiated protein peaks (P< 0.05). Genetic algorithm combining with SVM was utilized to establish the optimal diagnostic models for IBD and the optimal differential diagnostic model between CD and UC. The predictive effects of these models were evaluated by LOO cross validation method.Part two:serum proteome profiles from 30 CD patients,21 ITB patients and 30 healthy controls were analyzed using weak cationic magnetic beads combined with MALDI-TOF-MS technique to detect the differentially expressed proteins of serum samples. Three groups were compared:group of CD patients and healthy controls, group of ITB patients and healthy controls, group of CD patients and ITB patients. Wilcoxon rank sum test was used to screen the most differentiated protein peaks (P< 0.05). Genetic algorithm combining with SVM was utilized to establish the optimal diagnostic models for CD, ITB and the optimal differential diagnostic model between CD and ITB. The predictive effects of these models were evaluated by LOO cross validation method.A blind test with the diagnostic model (M/Z 4964,2272,2126,2900) of CD was performed to assess the diagnostic effect.Results:(1) Part one:There were 282 protein peaks differently expressed between group of CD patients and healthy controls,302 protein peaks differently expressed between group of UC patients and healthy controls,306 protein peaks differently expressed between group of CD patients and UC patients. Ten most discriminating peaks were screened out between three groups respectively (P<0.05) to establish and screen diagnostic models and differential diagnostic models. A diagnostic model consisting of four protein peaks (M/Z 4964,2272,2126,2900) can distinguish CD and healthy controls well, with a specificity of 93.8% and sensitivity of 81.4%. A diagnostic model comprising six protein peaks (M/Z 3030,5065,2360,3275,1945,3957) can distinguish UC and healthy controls well, with a specificity of 89.6% and sensitivity of 98.0%. A diagnostic model comprising four protein peaks (M/Z 2900,5338,2122, 1897) can distinguish CD and UC, with a specificity of 81.4% and sensitivity of 71.4%.(2) Part two:There were 236 protein peaks differently expressed between group of CD and healthy control,305 protein peaks differently expressed between group of ITB and healthy control,332 protein peaks differently expressed between group of CD and ITB. Ten most discriminating peaks were screened out between three groups respectively (P<0.05) to establish and screen diagnostic models and differential diagnostic models. A diagnostic model consisting of four protein peaks (M/Z 4964, 3029,2833,2900) can distinguish CD and healthy controls well, with a specificity of 96.7% and sensitivity of 96.7%. A diagnostic model comprising six protein peaks (M/Z 3030,2105,2545,4210) can distinguish ITB and healthy controls well, with a specificity of 93.3% and sensitivity of 95.2%. A diagnostic model comprising three potential biomarkers protein peaks (M/Z 4267,4223,1541) can distinguish CD and ITB, with a specificity of 76.2% and sensitivity of 90.0%.(3) A blind test with the diagnostic model (M/Z 4964,2272,2126,2900) of CD has a specificity of 96.7%, a sensitivity of 83.3% and an accuracy rate of 90%.Conclusion:(1) Part one:The differently expressed protein peaks analyzed by serum proteome with weak cationic magnetic beads combined MALDI-TOF-MS technique can effectively distinguish IBD patients and healthy controls. By genetic algorithm combining with SVM, the diagnostic model between CD and healthy controls consisting of four protein peaks (M/Z 4964,2272,2126,2900), the diagnostic model between UC and healthy controls comprising six protein peaks (M/Z 3030,5065, 2360,3275,1945,3957) and the differential diagnostic model between CD and UC comprising four protein peaks (M/Z 2900,5338,2122,1897) had high specificity and sensitivity and can help to the diagnosis of IBD and the differential diagnosis between subtypes of IBD.(2) Part two:The differently expressed protein peaks analyzed by serum proteome with weak cationic magnetic beads combined MALDI-TOF-MS technique can effectively distinguish CD patients, ITB patients and healthy controls. By genetic algorithm combining with SVM, the diagnostic model between CD and healthy controls consisting of four protein peaks (M/Z 4964,3029,2833,2900), the diagnostic model between ITB and healthy controls comprising four protein peaks (M/Z 3030,2105, 2545,4210) and the differential diagnostic model between CD and ITB comprising three potential biomarkers (M/Z 4267,4223,1541) had high specificity and sensitivity and can help to the diagnosis of CD, ITB and the differential diagnosis between CD and ITB.(3) The diagnostic model comprising four protein peaks (M/Z 4964,2272,2126, 2900) has a great advantage to diagnose CD. |
PDF Full Text Request