Benchmarking Classification Models for Cell Viability on Novel Cancer Image Datasets

Cancer classification based on multiple dimensions: SNV patterns

BACKGROUND

The occurrence of cancer is closely related to single nucleotide variants (SNVs). However, in DNA samples collected from patients with distinct cancers, SNVs are detected in different patterns. Therefore, it is an important task to select the appropriate method by which to classify cancer to the greatest extent of SNV patterns, which will aid in cancer diagnosis and treatment. In traditional studies, researchers combined each SNV with its neighboring nucleotides to form a trinucleotide. Mutation signatures for cancer classification were extracted from the patterns of the trinucleotides, but the SNV feature extraction in a single dimension may result in partial information loss and poor model performance.

RESULTS

In this study, we defined multidimensional SNV (M-SNV) features to classify cancer. M-SNV features considered first- and second-order neighboring nucleotides of one-dimensional SNVs and included six types of features. We validated the feasibility of M-SNV features using a dataset obtained from The Cancer Genome Atlas (TCGA) consisting of 2761 samples from 12 cancers. We performed preliminary screening of 562,321 DNA mutation sites in these samples. The remaining mutation sites were characterized by cancer type in six signatures. We found that the extracted features showed a similar distribution in the cluster center of the cancer type of the samples. After the preprocessing of raw data, samples were more focused on the cancer subtype distributions at the SNV level. We used KNN (k-nearest neighbors) to classify the extracted features and employed the leave-one-out cross to verify them. The accuracy of classifying is stable at approximately 97% and can reach 97.43% in the most optimal case. Furthermore, we found that the validated oncogenes in the loci of the features had the highest importance among the 8 cancers.

CONCLUSIONS

It is feasible to classify cancers by the distribution of features we defined. Moreover, our methodology has potential implications for the discovery of oncogenes.

Numerical learning of deep features from drug-exposed cell images to calculate IC50 without staining

To facilitate rapid determination of cellular viability caused by the inhibitory effect of drugs, numerical deep learning algorithms was used for unlabeled cell culture images captured by a light microscope as input. In this study, A549, HEK293, and NCI-H1975 cells were cultured, each of which have different molecular shapes and levels of drug responsiveness to doxorubicin (DOX). The microscopic images of these cells following exposure to various concentrations of DOX were trained with the measured value of cell viability using a colorimetric cell proliferation assay. Convolutional neural network (CNN) models for the study cells were constructed using augmented image data; the predicted cell viability using CNN models was compared to the cell viability measured by colorimetric assay. The linear relationship coefficient (r²) between measured and predicted cell viability was determined as 0.94–0.95 for the three cell types. In addition, the measured and predicted IC50 values were not statistically different. When drug responsiveness was estimated using allogenic models that were trained with a different cell type, the correlation coefficient decreased to 0.004085–0.8643. Our models could be applied to label-free cells to conduct rapid and large-scale tests while minimizing cost and labor, such as high-throughput screening for drug responsiveness.

Recent development of machine learning methods in sumoylation sites prediction

Sumoylation of proteins is an important reversible post-translational modification of proteins and mediates a variety of cellular processes. Sumo-modified proteins can change their subcellular localization, activity and stability. In addition, it also plays an important role in various cellular processes such as transcriptional regulation and signal transduction. The abnormal sumoylation is involved in many diseases, including neurodegeneration and immune-related diseases, as well as the development of cancer. Therefore, identification of the sumoylation site (SUMO site) is fundamental to understanding their molecular mechanisms and regulatory roles. In contrast to labor-intensive and costly experimental approaches, computational prediction of sumoylation sites in silico also attracted much attention for its accuracy, convenience and speed. At present, many computational prediction models have been used to identify SUMO sites, but these contents have not been comprehensively summarized and reviewed. Therefore, the research progress of relevant models is summarized and discussed in this paper. We will briefly summarize the development of bioinformatics methods on sumoylation site prediction. We will mainly focus on the benchmark dataset construction, feature extraction, machine learning method, published results and online tools. We hope the review will provide more help for wet-experimental scholars.

RNA-Associated Co-expression Network Identifies Novel Biomarkers for Digestive System Cancer

Cancers of the digestive system are malignant diseases. Our study focused on colon cancer, esophageal cancer (ESCC), rectal cancer, gastric cancer (GC), and rectosigmoid junction cancer to identify possible biomarkers for these diseases. The transcriptome data were downloaded from the TCGA database (The Cancer Genome Atlas Program), and a network was constructed using the WGCNA algorithm. Two significant modules were found, and coexpression networks were constructed. CytoHubba was used to identify hub genes of the two networks. GO analysis suggested that the network genes were involved in metabolic processes, biological regulation, and membrane and protein binding. KEGG analysis indicated that the significant pathways were the calcium signaling pathway, fatty acid biosynthesis, and pathways in cancer and insulin resistance. Some of the most significant hub genes were hsa-let-7b-3p, hsa-miR-378a-5p, hsa-miR-26a-5p, hsa-miR-382-5p, and hsa-miR-29b-2-5p and SECISBP2 L, NCOA1, HERC1, HIPK3, and MBNL1, respectively. These genes were predicted to be associated with the tumor prognostic reference for this patient population.

A Decision-Making Supporting Prediction Method for Breast Cancer Neoadjuvant Chemotherapy

Neoadjuvant chemotherapy (NAC) may increase the resection rate of breast cancer and shows promising effects on patient prognosis. It has become a necessary treatment choice and is widely used in the clinical setting. Benefitting from the clinical information obtained during NAC treatment, computational methods can improve decision-making by evaluating and predicting treatment responses using a multidisciplinary approach, as there are no uniformly accepted protocols for all institutions for adopting different treatment regiments. In this study, 166 Chinese breast cancer cases were collected from patients who received NAC treatment at the First Bethune Hospital of Jilin University. The Miller–Payne grading system was used to evaluate the treatment response. Four machine learning multiple classifiers were constructed to predict the treatment response against the 26 features extracted from the patients’ clinical data, including Random Forest (RF) model, Convolution Neural Network (CNN) model, Support Vector Machine (SVM) model, and Logistic Regression (LR) model, where the RF model achieved the best performance using our data. To allow a more general application, the models were reconstructed using only six selected features, and the RF model achieved the highest performance with 54.26% accuracy. This work can efficiently guide optimal treatment planning for breast cancer patients.

Prediction of Anticancer Peptides Using a Low-Dimensional Feature Model

Cancer is still a severe health problem globally. The therapy of cancer traditionally involves the use of radiotherapy or anticancer drugs to kill cancer cells, but these methods are quite expensive and have side effects, which will cause great harm to patients. With the find of anticancer peptides (ACPs), significant progress has been achieved in the therapy of tumors. Therefore, it is invaluable to accurately identify anticancer peptides. Although biochemical experiments can solve this work, this method is expensive and time-consuming. To promote the application of anticancer peptides in cancer therapy, machine learning can be used to recognize anticancer peptides by extracting the feature vectors of anticancer peptides. Nevertheless, poor performance usually be found in training the machine learning model to utilizing high-dimensional features in practice. In order to solve the above job, this paper put forward a 19-dimensional feature model based on anticancer peptide sequences, which has lower dimensionality and better performance than some existing methods. In addition, this paper also separated a model with a low number of dimensions and acceptable performance. The few features identified in this study may represent the important features of anticancer peptides.

Prediction of K562 Cells Functional Inhibitors Based on Machine Learning Approaches

BACKGROUND

β thalassemia is a common monogenic genetic disease that is very harmful to human health. The disease arises is due to the deletion of or defects in β-globin, which reduces synthesis of the β-globin chain, resulting in a relatively excess number of α-chains. The formation of inclusion bodies deposited on the cell membrane causes a decrease in the ability of red blood cells to deform and a group of hereditary haemolytic diseases caused by massive destruction in the spleen.

METHODS

In this work, machine learning algorithms were employed to build a prediction model for inhibitors against K562 based on 117 inhibitors and 190 non-inhibitors.

RESULTS

The overall accuracy (ACC) of a 10-fold cross-validation test and an independent set test using Adaboost were 83.1% and 78.0%, respectively, surpassing Bayes Net, Random Forest, Random Tree, C4.5, SVM, KNN and Bagging.

CONCLUSION

This study indicated that Adaboost could be applied to build a learning model in the prediction of inhibitors against K526 cells.