Logistic regression with adaptive sparse group lasso penalty and its application in acute leukemia diagnosis

Prediction of g-C3N4-based photocatalysts in tetracycline degradation based on machine learning

Investigating the effects of g-C3N4-based photocatalysts on experimental parameters during tetracycline (TC) degradation can be helpful in discovering the optimal parameter combinations to improve the degradation efficiencies in general. Machine learning methods can avoid the problems of high cost, time-consuming and possible instrumental errors in experimental methods, which have been proven to be an effective alternative for evaluating the entire experimental process. Eight typical machine learning models were explored for their effectiveness in predicting the TC degradation efficiencies of g-C3N4 based photocatalysts. XGBoost (XGB) was the most reliable model with R2, RMSE and MAE values of 0.985, 4.167 and 2.900, respectively. In addition, XGB's feature importance and SHAP method were used to rank the importance of features to provide interpretability to the results. This study provided a new idea for developing g-C3N4-based photocatalysts for TC degradation and intelligent algorithms for predicting the photocatalytic activity of g-C3N4-based photocatalysts.

Bi-level gene selection of cancer by combining clustering and sparse learning

The diagnosis of cancer based on gene expression profile data has attracted extensive attention in the field of biomedical science. This type of data usually has the characteristics of high dimensionality and noise. In this paper, a hybrid gene selection method based on clustering and sparse learning is proposed to choose the key genes with high precision. We first propose a filter method, which combines the k-means clustering algorithm and signal-to-noise ratio ranking method, and then, a weighted gene co-expression network has been applied to the reduced data set to identify modules corresponding to biological pathways. Moreover, we choose the key genes by using group bridge and sparse group lasso as wrapper methods. Finally, we conduct some numerical experiments on six cancer datasets. The numerical results show that our proposed method has achieved good performance in gene selection and cancer classification.

Nomogram prediction model of postoperative pneumonia in patients with lung cancer: A retrospective cohort study

Background

The prediction model of postoperative pneumonia (POP) after lung cancer surgery is still scarce.

Methods

Retrospective analysis of patients with lung cancer who underwent surgery at The Fourth Hospital of Hebei Medical University from September 2019 to March 2020 was performed. All patients were randomly divided into two groups, training cohort and validation cohort at the ratio of 7:3. The nomogram was formulated based on the results of multivariable logistic regression analysis and clinically important factors associated with POP. Concordance index (C-index), receiver operating characteristic (ROC) curve, calibration curve, Hosmer-Lemeshow goodness-of-fit test and decision curve analysis (DCA) were used to evaluate the predictive performance of the nomogram.

Results

A total of 1252 patients with lung cancer was enrolled, including 877 cases in the training cohort and 375 cases in the validation cohort. POP was found in 201 of 877 patients (22.9%) and 89 of 375 patients (23.7%) in the training and validation cohorts, respectively. The model consisted of six variables, including smoking, diabetes mellitus, history of preoperative chemotherapy, thoracotomy, ASA grade and surgery time. The C-index from AUC was 0.717 (95%CI:0.677-0.758) in the training cohort and 0.726 (95%CI:0.661-0.790) in the validation cohort. The calibration curves showed the model had good agreement. The result of DCA showed that the model had good clinical benefits.

Conclusion

This proposed nomogram could predict the risk of POP in patients with lung cancer surgery in advance, which can help clinician make reasonable preventive and treatment measures.

Development and validation of a novel immune-related prognostic signature in lung squamous cell carcinoma patients

Lung Squamous Cell Carcinoma (LUSC) is an aggressive malignancy with limited therapeutic options. The response to immune therapy is a determining factor for the prognosis of LUSC patients. This study aimed to develop a reliable immune-related prognostic signature in LUSC. We extracted gene expression and clinical data of LUSC from The Cancer Genome Atlas (TCGA). A total of 502 patients enrolled and were divided into respond and non-responder groups by the TIDE algorithm. The CIBERSORT algorithm and the LM22 gene signature were used to analyze the distribution of immune cells in LUSC. Efficacy and response strength of immunotherapy are calculated by the tumor mutation burden (TMB) and ESTIMATE Score. Differentially expressed genes (DEGs) between the two groups were analyzed. The differential expression genes related to overall survival were pointed as hub DEGs, and a prognostic signature was constructed with lasso regression analysis. LUSC patients were divided into responder and non-responder groups based on the response to immunotherapy. The distribution of immune cells was significantly different between the two groups. Forty-four DGEs were considered as overall survival-related genes. A prognostic signature was constructed, consisting of 11 hub-DGEs, including MMP20, C18orf26, CASP14, FAM71E2, OPN4, CGB5, DIRC1, C9orf11, SPATA8, C9orf144B, and ZCCHC5. The signature can accurately distinguish LUSC patients into high and low-risk groups. Moreover, the high-risk group had a shorter survival time than the low-risk group. The area under the ROC curve was 0.67. The multivariate Cox regression showed that the risk score calculated by the constructed signature was an independent prognostic predictor for LUSC patients. In short, we established a novel immune-related prognostic signature in LUCS, which has significant sensitivity and accuracy in predicting the prognosis of patients. Our research can guide the evaluation of the prognosis of LUSC patients in clinical, and the discovered immune-related genes can provide a theoretical basis for the discovery of new therapeutic targets.

Identification of miRNA biomarkers for breast cancer by combining ensemble regularized multinomial logistic regression and Cox regression

BACKGROUND

Breast cancer is one of the most common cancers in women. It is necessary to classify breast cancer subtypes because different subtypes need specific treatment. Identifying biomarkers and classifying breast cancer subtypes is essential for developing appropriate treatment methods for patients. MiRNAs can be easily detected in tumor biopsy and play an inhibitory or promoting role in breast cancer, which are considered promising biomarkers for distinguishing subtypes.

RESULTS

A new method combing ensemble regularized multinomial logistic regression and Cox regression was proposed for identifying miRNA biomarkers in breast cancer. After adopting stratified sampling and bootstrap sampling, the most suitable sample subset for miRNA feature screening was determined via ensemble 100 regularized multinomial logistic regression models. 124 miRNAs that participated in the classification of at least 3 subtypes and appeared at least 50 times in 100 integrations were screened as features. 22 miRNAs from the proposed feature set were further identified as the biomarkers for breast cancer by using Cox regression based on survival analysis. The accuracy of 5 methods on the proposed feature set was significantly higher than on the other two feature sets. The results of 7 biological analyses illustrated the rationality of the identified biomarkers.

CONCLUSIONS

The screened features can better distinguish breast cancer subtypes. Notably, the genes and proteins related to the proposed 22 miRNAs were considered oncogenes or inhibitors of breast cancer. 9 of the 22 miRNAs have been proved to be markers of breast cancer. Therefore, our results can be considered in future related research.

Ensemble classification based signature discovery for cancer diagnosis in RNA expression profiles across different platforms

Molecular signatures have been excessively reported for diagnosis of many cancers during the last 20 years. However, false-positive signatures are always found using statistical methods or machine learning approaches, and that makes subsequent biological experiments fail. Therefore, signature discovery has gradually become a non-mainstream work in bioinformatics. Actually, there are three critical weaknesses that make the identified signature unreliable. First of all, a signature is wrongly thought to be a gene set, each component of which keeps differential expressions between or among sample groups. Second, there may be many false-positive genes expressed differentially found, even if samples derived from cancer or normal group can be separated in one-dimensional space. Third, cross-platform validation results of a discovered signature are always poor. In order to solve these problems, we propose a new feature selection framework based on ensemble classification to discover signatures for cancer diagnosis. Meanwhile, a procedure for data transform among different expression profiles across different platforms is also designed. Signatures are found on simulation and real data representing different carcinomas across different platforms. Besides, false positives are suppressed. The experimental results demonstrate the effectiveness of our method.