scholarly journals Tumor marker based survival analysis for patients with pseudomyxoma peritonei of appendiceal origin: A retrospective cohort study

2020 ◽  
Author(s):  
Mingjian Bai ◽  
Shilong Wang ◽  
Ruiqing Ma ◽  
Ying Cai ◽  
Yiyan Lu ◽  
...  
Keyword(s):  
Prediction Model ◽  
Pseudomyxoma Peritonei ◽  
Roc Analysis ◽  
Prediction Models ◽  
Radical Surgery ◽  
Univariate Analysis ◽  
Barthel Index Score ◽  
Discrimination Ability ◽  
First Time ◽  
D Dimer

Abstract Background Pseudomyxoma peritonei (PMP) is a rare disease, the prognosis of overall survival (OS) is affected by many factors, present study aim to screen independent prediction indicators for PMP and establish prediction model for OS rates in PMP.Methods 119 PMP patients received cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) in our center for the first time were included between 01/06/2013 and 22/11/2019 . The log-rank test was used to compare the OS rate among groups, subsequently, variables with P<0.10 were subjected to multivariate Cox model for screening independent prediction indicators. Finally, the prediction models for OS in PMP will be established.Results Univariate analysis showed that Barthel Index Score, albumin, D-dimer, CEA, CA125, CA19-9, CA724, CA242, PCI, degree of radical surgery, histopathological grade were significant predictors for OS in PMP. At multivariate analysis, sex, D-dimer, CA125, CA19-9, and degree of radical surgery were independently associated with OS rate in PMP. ROC analysis was performed to calculate discrimination ability of prediction model and the area under curves (AUC) was 0.902 (95%CI: 0.823-0.954). Finally, nomogram was plotted by the independent predictive factors for PMP.Conclusions Several factors (sex, degree of radical surgery, D-dimer, preoperative CA125 and CA19-9) have independent prognostic value for survival in PMP, the tumor based prediction model has better prediction value, more researches are need to verify and improve the prediction model.

scholarly journals Tumor marker based survival analysis for patients with pseudomyxoma peritonei of appendiceal origin: A retrospective cohort study

2020 ◽  
Author(s):  
Mingjian Bai ◽  
Shilong Wang ◽  
Ruiqing Ma ◽  
Ying Cai ◽  
Yiyan Lu ◽  
...  
Keyword(s):  
Prediction Model ◽  
Pseudomyxoma Peritonei ◽  
Prediction Models ◽  
Radical Surgery ◽  
Cox Model ◽  
Univariate Analysis ◽  
Rank Test ◽  
Internal Validation ◽  
Discrimination Ability ◽  
D Dimer

Abstract Background Pseudomyxoma peritonei (PMP) is a rare disease, the prognosis of overall survival (OS) is affected by many factors, present study aim to screen independent prediction indicators and establish a nomogram for PMP. Methods 119 PMP patients received cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) in our center for the first time were included between 01/06/2013 and 22/11/2019 . The log-rank test was used to compare the OS rate among groups, subsequently, variables with P<0.10 were subjected to multivariate Cox model for screening independent prediction indicators. Finally, the nomogram prediction models will be established. Results Univariate analysis showed that Barthel Index Score, Albumin, D-Dimer, CEA, CA125, CA19-9, CA724, CA242, PCI, degree of radical surgery, histopathological grade were significant predictors for OS in PMP. At multivariate analysis, Sex, D-Dimer, CA125, CA19-9, PCI, and degree of radical surgery were independently associated with OS rate in PMP. A nomogram was plotted based on the independent predictive factors for PMP and undergone internal validation, ROC analysis was performed to calculate discrimination ability of prediction model, the area under curves (AUC) was 0.880 (95% CI : 0.806- 0.933). Conclusions Several factors (Sex, D-Dimer, CA125, CA19-9, PCI, and degree of radical surgery) have independent prognostic value for survival in PMP, the tumor based prediction model has a better prediction value, more researches are need to verify and improve the prediction model.

scholarly journals Nomogram to predict overall survival for patients with pseudomyxoma peritonei of appendiceal origin: A retrospective cohort study

2020 ◽  
Author(s):  
Mingjian Bai ◽  
Shilong Wang ◽  
Ruiqing Ma ◽  
Ying Cai ◽  
Yiyan Lu ◽  
...  
Keyword(s):  
Overall Survival ◽  
Pseudomyxoma Peritonei ◽  
Roc Analysis ◽  
Prediction Models ◽  
Radical Surgery ◽  
Individual Risk ◽  
Good Discrimination ◽  
Internal Validation ◽  
Discrimination Ability ◽  
First Time

Abstract Background Pseudomyxoma peritonei (PMP) is a rare disease, the prognosis of overall survival (OS) is affected by many factors, present study aim to define independent prediction indicators and establish a nomogram for PMP patients.Methods 119 PMP patients received cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) in our center for the first time were included between 01/06/2013 and 22/11/2019 . The log-rank test was used to compare the OS rate among groups, subsequently, variables with P<0.10 were subjected to multivariate Cox model for defining independent prediction indicators. Finally, the nomogram prediction models will be established and for internal validation.Results Multivariate analysis showed Sex, D-Dimer, CA125, CA19-9, PCI, and degree of radical surgery were independently associated with OS in PMP patients. A nomogram was plotted based on the independent predictive factors and undergone internal validation, ROC analysis was performed to calculate discrimination ability of the nomogram, the C-index was 0.880 (95%CI: 0.806- 0.933) and calibration plots showed good performance. Conclusions Six independent prognostic factor for predicting survival in PMP patients were difined, the nomogram has a good discrimination ability for individual risk predition, more researches are needed to verify and improve the prediction model.

scholarly journals Identification and Characterization of Circular RNA as a Novel Regulator and Biomarker in Preterm Birth

2020 ◽  
Vol 8 ◽  
Author(s):  
Yuxin Ran ◽  
Nanlin Yin ◽  
Dongni Huang ◽  
Yangyu Zhao ◽  
Jing Yang ◽  
...  
Keyword(s):  
Preterm Birth ◽  
Roc Analysis ◽  
Underlying Mechanism ◽  
Circular Rna ◽  
Diagnosis And Treatment ◽  
Rna Seq ◽  
Discrimination Ability ◽  
First Time ◽  
Identification And Characterization

Preterm birth (PTB), as the leading cause of neonatal death, is a severe threat to maternal–fetal health. The diagnosis and treatment of PTB are difficult as its underlying mechanism still unknown. Circular RNA (circRNA) is an emerging molecule that plays an essential role in the pathological processes of various diseases. However, it is still unclear whether circRNAs are abnormal or involves in the PTB pathology. In this study, we analyzed RNA-seq data of peripheral blood from preterm and term pregnant women and verified with microarray data. There were 211 circRNA expression disorders in PTB, of which 68 increased and 143 decreased. Bioinformatics analysis revealed that the top 20 circRNAs competitively bind 68 miRNAs, thereby regulating 622 mRNAs mainly related to immunity, inflammation, and nerve activity, which may ultimately contribute to the occurrence of PTB. Moreover, 6 regulatory pairs, including hsa-MORC3_0001–hsa-miR-1248–CHRM2 were the core parts of this mechanism network, which might be therapeutic targets for PTB. Besides, ROC analysis indicated that hsa-ANKFY1_0025, hsa-FAM13B_0019, and hsa-NUSAP1_0010 (AUC = 0.7138, 0.9589, 1.000) have an excellent discrimination ability for PTB. Taken together, we explored for the first time the circRNA expression profile of PTB, and preliminarily analyzed its regulatory mechanism and predictive value for PTB, thus bringing new light to the diagnosis and treatment of PTB.

scholarly journals Development and external validation of a prognostic multivariable model on admission for hospitalized patients with COVID-19

2020 ◽  
Author(s):  
Jianfeng Xie ◽  
Daniel Hungerford ◽  
Hui Chen ◽  
Simon T Abrams ◽  
Shusheng Li ◽  
...  
Keyword(s):  
Lactate Dehydrogenase ◽  
Prediction Model ◽  
Lymphocyte Count ◽  
Prediction Models ◽  
External Validation ◽  
Model Development ◽  
Added Value ◽  
Clinical Prediction ◽  
Healthcare Resources ◽  
D Dimer

SummaryBackgroundCOVID-19 pandemic has developed rapidly and the ability to stratify the most vulnerable patients is vital. However, routinely used severity scoring systems are often low on diagnosis, even in non-survivors. Therefore, clinical prediction models for mortality are urgently required.MethodsWe developed and internally validated a multivariable logistic regression model to predict inpatient mortality in COVID-19 positive patients using data collected retrospectively from Tongji Hospital, Wuhan (299 patients). External validation was conducted using a retrospective cohort from Jinyintan Hospital, Wuhan (145 patients). Nine variables commonly measured in these acute settings were considered for model development, including age, biomarkers and comorbidities. Backwards stepwise selection and bootstrap resampling were used for model development and internal validation. We assessed discrimination via the C statistic, and calibration using calibration-in-the-large, calibration slopes and plots.FindingsThe final model included age, lymphocyte count, lactate dehydrogenase and SpO2 as independent predictors of mortality. Discrimination of the model was excellent in both internal (c=0·89) and external (c=0·98) validation. Internal calibration was excellent (calibration slope=1). External validation showed some over-prediction of risk in low-risk individuals and under-prediction of risk in high-risk individuals prior to recalibration. Recalibration of the intercept and slope led to excellent performance of the model in independent data.InterpretationCOVID-19 is a new disease and behaves differently from common critical illnesses. This study provides a new prediction model to identify patients with lethal COVID-19. Its practical reliance on commonly available parameters should improve usage of limited healthcare resources and patient survival rate.FundingThis study was supported by following funding: Key Research and Development Plan of Jiangsu Province (BE2018743 and BE2019749), National Institute for Health Research (NIHR) (PDF-2018-11-ST2-006), British Heart Foundation (BHF) (PG/16/65/32313) and Liverpool University Hospitals NHS Foundation Trust in UK.Research in contextEvidence before this studySince the outbreak of COVID-19, there has been a pressing need for development of a prognostic tool that is easy for clinicians to use. Recently, a Lancet publication showed that in a cohort of 191 patients with COVID-19, age, SOFA score and D-dimer measurements were associated with mortality. No other publication involving prognostic factors or models has been identified to date.Added value of this studyIn our cohorts of 444 patients from two hospitals, SOFA scores were low in the majority of patients on admission. The relevance of D-dimer could not be verified, as it is not included in routine laboratory tests. In this study, we have established a multivariable clinical prediction model using a development cohort of 299 patients from one hospital. After backwards selection, four variables, including age, lymphocyte count, lactate dehydrogenase and SpO2 remained in the model to predict mortality. This has been validated internally and externally with a cohort of 145 patients from a different hospital. Discrimination of the model was excellent in both internal (c=0·89) and external (c=0·98) validation. Calibration plots showed excellent agreement between predicted and observed probabilities of mortality after recalibration of the model to account for underlying differences in the risk profile of the datasets. This demonstrated that the model is able to make reliable predictions in patients from different hospitals. In addition, these variables agree with pathological mechanisms and the model is easy to use in all types of clinical settings.Implication of all the available evidenceAfter further external validation in different countries the model will enable better risk stratification and more targeted management of patients with COVID-19. With the nomogram, this model that is based on readily available parameters can help clinicians to stratify COVID-19 patients on diagnosis to use limited healthcare resources effectively and improve patient outcome.

scholarly journals Development and validation of postoperative predictive recurrence nomograms in patients with persistent atrial fibrillation: a retrospective cohort study

2021 ◽  
Author(s):  
Cong-ying Deng ◽  
Ling Sun ◽  
Yuan Ji
Keyword(s):  
Risk Factors ◽  
Atrial Fibrillation ◽  
Prediction Model ◽  
Linear Prediction ◽  
Preoperative Evaluation ◽  
Univariate Analysis ◽  
Persistent Atrial Fibrillation ◽  
Risk Of Recurrence ◽  
Discrimination Ability ◽  
Calibration Curves

Background： Patients with persistent atrial fibrillation（PsAF） still have a higher risk of recurrence after catheter radiofrequency ablation. Nevertheless, effective recurrence forecast tools have not been established for these patients. Thus, this research aimed to develop and validate an easy-to-use linear prediction model for predicting postoperative recurrence in patients with PsAF. Methods： We conducted a single-center, retrospective, observational study of patients with PsAF admitted to our hospital from June 2013 to June 2021. Univariate analysis was used to screen risk factors, then we used multivariate regression analysis to evaluate the independence of each risk factor and construct a combined prediction model which incorporated into a nomogram, finally, we took the receiver operating characteristic (ROC) curve to predict the value of nomogram model. Additionally, the calibration curves and decision curve analysis (DCA) were also performed to assess the clinical utility of the nomogram. Results： A total of 209 subjects were included in the study and 42 (20.10%) were followed up to September 2021 for recurrent AF.  Duration of AF, Left atrial diameter(LAD), BMI, CKMB, and alcohol consumption were independent risk factors (P < 0.05),  these variables were integrated into the nomogram model, and the area under the curve （AUC） was 0.895, the sensitivity was 93.3%, and the specificity was 71.4%, indicating that the model had an excellent predictive performance.  The C-index of the predictive nomogram model was 0.906. The calibration curves and DCA results manifested that the Model had a splendid predictive correction and discrimination ability. Conclusion： This simple and innovative clinical nomogram (that any clinician can use in the daily clinic) can help evaluate the risk of recurrence after catheter ablation in PsAF, facilitate preoperative evaluation as well as the postoperative follow-up, and may also help generate personalized therapeutic strategies.

Predicting Severe Renal and Gastrointestinal Involvement in Childhood Immunoglobulin A Vasculitis with Routine Laboratory Parameters

Dermatology ◽  
2021 ◽  
pp. 1-8
Author(s):  
Zexing Song ◽  
Yingli Nie ◽  
Liu Yang ◽  
Juan Tao
Keyword(s):  
Prediction Model ◽  
Prediction Models ◽  
Renal Involvement ◽  
Immunoglobulin A ◽  
Prognostic Information ◽  
Cutaneous Manifestations ◽  
Laboratory Parameters ◽  
Wbc Count ◽  
Immunoglobulin A Vasculitis ◽  
D Dimer

<b><i>Background:</i></b> Immunoglobulin A vasculitis (IgAV) is the most common vasculitis in children. Although childhood IgAV is generally considered as a self-limited disease, progressive course and poor prognosis could occur in some cases which mostly result from severe renal involvement and gastrointestinal (GI) involvement. <b><i>Methods:</i></b> We performed a retrospective study of pediatric patients diagnosed as IgAV in our institution from 2016 to 2019. Patients were divided into groups based on the occurrence and severity of GI and renal involvement. Analysis of variance (ANOVA) and Kruskal-Wallis test were used to compare results of laboratory parameters among groups and prediction models were built by using logistic regression analysis. <b><i>Results:</i></b> A total of 286 patients were enrolled. GI involvement occurred in 148 (51.7%) patients, 30 (20.3%) of which were severe cases. Renal involvement developed in 120 (42.0%) patients, 22 (18.3%) of which were severe cases. Compared with patients with only cutaneous manifestations, white blood cell (WBC) count, neutrophil-to-lymphocyte ratio (NLR), and D-dimer levels were higher in those with GI involvement, and D-dimer level was found to be positively associated with severity. Increased NLR and lower complement 3 (C3) were found in patients with renal involvement, but only C3 was relevant in distinguishing moderate and severe cases. The prediction model for severe renal involvement was: Logit (P) = 6.820 + 0.270 (age) + 0.508 (NLR) − 16.130 (C3), with an AUC of 0.914. The prediction model for severe GI involvement was: Logit (P) = −5.459 + 0.005 (WBC) + 1.355 (D-dimer) – 0.020 (NLR), with an AUC of 0.849. <b><i>Conclusion:</i></b> Our data suggest C3 to be an exclusive predictor for severe renal involvement and D-dimer level to be positively associated with the severity of GI involvement. Prediction models consisting of the above parameters were built for obtaining prognostic information in the early phase of IgAV.

FRI0129 DEVELOPMENT OF A PREDICTION MODEL FOR MAXIMUM METHOTREXATE (MTX) DOSE WITHOUT HEPATOTOXICITY USING AN INDEX OF ERYTHROCYTE MTX-POLYGLUTAMATE (MTXPG) LEVELS SPECULATED BY CLINICAL AND GENETIC MARKERS

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 646-647
Author(s):  
S. Kumagai ◽  
S. Takahashi ◽  
M. Takahashi ◽  
T. Saito ◽  
K. Yoshida ◽  
...  
Keyword(s):  
Prediction Model ◽  
Genetic Markers ◽  
Prediction Models ◽  
Univariate Analysis ◽  
Clinical Pathology ◽  
Clinical Prediction Models ◽  
Clinical Model ◽  
Upper Limits ◽  
Calculated Dose ◽  
Rbc Count

Background:MTX is transported into cells and retained long after polyglutamation. MTXPG level can predict response and possibly adverse effects of MTX. We reported erythrocyte MTXPG concentrations efficiently discriminated patients with and without hepatotoxicity1. We also developed genetic and clinical prediction models for efficacy and hepatotoxicity of MTX2. In the present study, we firstly investigated the effects of clinical and secondly genetic variables on the concentration of total MTXPG and determined oral maximum MTX dose without hepatotoxicity using these variables.Objectives:To develop a prediction model for maximum MTX dose without hepatotoxicity.Methods:Concentrations of erythrocyte MTX-PG (PG1 to PG4) were detected by LC-MS/MS and calculated total MTXPG as sum of them. MTX-PGn levels were measured in 265 RA patients including 40 patients with elevated AST or ALT (≥ 60 U/L; 1.5 times of upper limits) and the 6 SNPs of 6 gens related to MTXPG metabolism were identified by RT-PCR.Results:Total concentrations of MTXPG were 141.3 ± 86.5 and 87.6 ± 47.8 nmol/L (mean±SD) in 40 RA patients with hepatotoxicity and 225 patients without, respectively (p<0.0001). By ROC analysis, the two groups were most efficiently discriminated with cutoff concentration of 100.0 nmol/L (AUC 0.731). Next, genetic and clinical model to speculate the MTXPG concentration was established by multivariate analysis using 4 clinical and 3 genetic variables which were selected from 20 clinical and 6 genetic variables by univariate analysis (p<0.1). Finally, a speculation model for MTXPG concentration by 4 clinical variables (MTX dose, BMI, RBC count, and creatinine) and one genetic variable (GGH c.452C>T) was developed (Figure). When MTXPG concentration of 100 nmol/L was applied to the model, maximum MTX dose without hepatotoxicity was calculated for each patient asMTX dose (mg) = {100 (MTXPG) – 96 + 1.7*BMI + 28*RBC - 120*creatinine - 19.3*GGH(C/T)} / 7.7. Real dose of oral MTX exceeded the calculated dose in 23 of 40 patients (57.5%) with hepatotoxicity, whereas it exceeded in 95 of 223 patients (42.6%) without hepatotoxicity (OR 1.82, p=0.081).Conclusion:Maximum MTX dose without hepatotoxicity was speculated by several clinical and genetic markers without measurement of erythrocyte MTX-PG concentrations.References:[1]Takahashi M, et al: Clinical Pathology (Rinsho Byori), 67:433-442, 2019.[2]Onishi A, et al: The Pharmacogenomics J, doi.org/10.1038/s41397-019-0134-9, 2019Disclosure of Interests:Shunichi Kumagai Grant/research support from: Astellas, Chugai, Mitsubishi Tanabe Co.Ltds, Consultant of: Sysmex Co.Ltd, Speakers bureau: many companies, Soshi Takahashi: None declared, Miho Takahashi: None declared, Toshiharu Saito: None declared, Katsuyuki Yoshida: None declared, Motoko Katayama: None declared, Saki Mukohara: None declared, Norihiko Amano: None declared, Akira Onishi Speakers bureau: AO received a speaker fee from Chugai, Ono Pharmaceutical, Eli Lilly, Mitsubishi-Tanabe, Asahi-Kasei, and Takeda, Masakazu Shinohara: None declared, Saori Hatachi: None declared

4 Pattern Recognition for Healing Burns to Compute Evidence: Space Syntax and Machine Learning Analysis of Burns Center Karachi

2013 ◽  
Vol 1 (1) ◽  
pp. 13
Author(s):  
Javaria Manzoor Shaikh ◽  
JaeSeung Park
Keyword(s):  
Pattern Recognition ◽  
Prediction Models ◽  
Uv Light ◽  
Healing Process ◽  
Space Syntax ◽  
Depth Map ◽  
Routine Tasks ◽  
Hyper Plane ◽  
First Time ◽  
Learning Analysis

Usually elongated hospitalization is experienced byBurn patients, and the precise forecast of the placement of patientaccording to the healing acceleration has significant consequenceon healthcare supply administration. Substantial amount ofevidence suggest that sun light is essential to burns healing andcould be exceptionally beneficial for burned patients andworkforce in healthcare building. Satisfactory UV sunlight isfundamental for a calculated amount of burn to heal; this delicaterather complex matrix is achieved by applying patternclassification for the first time on the space syntax map of the floorplan and Browder chart of the burned patient. On the basis of thedata determined from this specific healthcare learning technique,nurse can decide the location of the patient on the floor plan, hencepatient safety first is the priority in the routine tasks by staff inhealthcare settings. Whereas insufficient UV light and vitamin Dcan retard healing process, hence this experiment focuses onmachine learning design in which pattern recognition andtechnology supports patient safety as our primary goal. In thisexperiment we lowered the adverse events from 2012- 2013, andnearly missed errors and prevented medical deaths up to 50%lower, as compared to the data of 2005- 2012 before this techniquewas incorporated.In this research paper, three distinctive phases of clinicalsituations are considered—primarily: admission, secondly: acute,and tertiary: post-treatment according to the burn pattern andhealing rate—and be validated by capable AI- origin forecastingtechniques to hypothesis placement prediction models for eachclinical stage with varying percentage of burn i.e. superficialwound, partial thickness or full thickness deep burn. Conclusivelywe proved that the depth of burn is directly proportionate to thedepth of patient’s placement in terms of window distance. Ourfindings support the hypothesis that the windowed wall is mosthealing wall, here fundamental suggestion is support vectormachines: which is most advantageous hyper plane for linearlydivisible patterns for the burns depth as well as the depth map isused.

EFFICIENCY OF ADJUVANT XELOX CHEMOTHERAPY FOR PATIENTS WITH RESECTABLE GASTRIC IN RUSSIA (SINGLE INSTITUTION RETROSPECTIVE STUDY)

2019 ◽  
Vol 65 (2) ◽  
pp. 256-262
Author(s):  
Ivan Stilidi ◽  
Sergey Nered ◽  
Aleksey Kalinin ◽  
Olesya Rossomakhina ◽  
Anton Barchuk
Keyword(s):  
Gastric Cancer ◽  
Adjuvant Chemotherapy ◽  
Radical Surgery ◽  
Univariate Analysis ◽  
European Population ◽  
Single Institution ◽  
Regional Lymph Nodes ◽  
Lymph Nodes Metastasis ◽  
Retrospective Assessment ◽  
Resectable Gastric Cancer

Introduction. The effectiveness of the Asian regimen of adjuvant chemotherapy in patients with gastric cancer in the European population remains unclear. The aim of our study was a retrospective assessment of adjuvant chemotherapy (XELOX regimen) after radical surgery (R0) on overall survival. Methods. Database of pts with resectable gastric cancer with stage >pT3 and/or pN+ and M0, who were operated (R0) at single oncological institution during 2007-2017 was reviewed. In univariate and multivariate analyzes were included demographic characteristics, type of tumor according to Lauren, stage, type of treatment and others. Results. 396 pts were identified and 286 were available for analysis.106 (37%) pts received at least one cycle of adjuvant chemotherapy. In univariate analysis, 5OS rate was 64% [95% Cl, 52-80] и 56% [95% Cl, 48-64; p=0,21] in patients received adjuvant chemotherapy and only surgical treatment. After stratifying patients depending on the regional lymph nodes metastasis, 5OS rate in pts with pN1-3 was 69% [95% CI, 57-85] vs 47% [95% CI, 39-58; p = 0,01], respectively...

Bioactivity Prediction Based on Matched Molecular Pair and Matched Molecular Series Methods

2020 ◽  
Vol 26 (33) ◽  
pp. 4195-4205
Author(s):  
Xiaoyu Ding ◽  
Chen Cui ◽  
Dingyan Wang ◽  
Jihui Zhao ◽  
Mingyue Zheng ◽  
...  
Keyword(s):  
Prediction Model ◽  
Large Scale ◽  
Prediction Models ◽  
Predictive Accuracy ◽  
Lead Optimization ◽  
Consensus Method ◽  
Molecular Pair ◽  
Bioactivity Prediction ◽  
Compound Synthesis ◽  
Consensus Modeling

Background: Enhancing a compound’s biological activity is the central task for lead optimization in small molecules drug discovery. However, it is laborious to perform many iterative rounds of compound synthesis and bioactivity tests. To address the issue, it is highly demanding to develop high quality in silico bioactivity prediction approaches, to prioritize such more active compound derivatives and reduce the trial-and-error process. Methods: Two kinds of bioactivity prediction models based on a large-scale structure-activity relationship (SAR) database were constructed. The first one is based on the similarity of substituents and realized by matched molecular pair analysis, including SA, SA_BR, SR, and SR_BR. The second one is based on SAR transferability and realized by matched molecular series analysis, including Single MMS pair, Full MMS series, and Multi single MMS pairs. Moreover, we also defined the application domain of models by using the distance-based threshold. Results: Among seven individual models, Multi single MMS pairs bioactivity prediction model showed the best performance (R2 = 0.828, MAE = 0.406, RMSE = 0.591), and the baseline model (SA) produced the most lower prediction accuracy (R2 = 0.798, MAE = 0.446, RMSE = 0.637). The predictive accuracy could further be improved by consensus modeling (R2 = 0.842, MAE = 0.397 and RMSE = 0.563). Conclusion: An accurate prediction model for bioactivity was built with a consensus method, which was superior to all individual models. Our model should be a valuable tool for lead optimization.

Sign in / Sign up

Export Citation Format

Share Document