Performance Comparisons of AlexNet and GoogLeNet in Cell Growth Inhibition IC50 Prediction

Lee, Yeeun; Nam, Seungyoon

doi:10.3390/ijms22147721

Cited by 13 publications

(8 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are several reports to predict IC50 of anti-cancer drugs using genomic profiles and drug fingerprints 18 , 19 . In these studies, genomic profiles of various cancer cell lines and fingerprints of drugs were trained to categorize IC50 levels in three steps, high responsiveness (class 0), intermediate responsiveness (class 1), and low responsiveness (class 2).…”

Section: Resultsmentioning

confidence: 99%

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

Cho

Choi

Kim

et al. 2022

Sci Rep

View full text Add to dashboard Cite

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.

show abstract

Section: Resultsmentioning

confidence: 99%

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

Cho

Choi

Kim

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Our results suggest that patients in the low-risk group have lower TIDE and higher IPS score; therefore, patients in the low-risk group were more sensitive to immunotherapy. IC50 is an important index for predicting chemosensitivity ( 52 ). Based on the IC50 results, we found significant differences in the sensitivity of patients in the high- and low-risk groups to 33 common chemotherapy drugs.…”

Section: Discussionmentioning

confidence: 99%

Development of a novel hypoxia-immune–related LncRNA risk signature for predicting the prognosis and immunotherapy response of colorectal cancer

et al. 2022

View full text Add to dashboard Cite

BackgroundColorectal cancer (CRC) is one of the most common digestive system tumors worldwide. Hypoxia and immunity are closely related in CRC; however, the role of hypoxia-immune–related lncRNAs in CRC prognosis is unknown.MethodsData used in the current study were sourced from the Gene Expression Omnibus and The Cancer Genome Atlas (TCGA) databases. CRC patients were divided into low- and high-hypoxia groups using the single-sample gene set enrichment analysis (ssGSEA) algorithm and into low- and high-immune groups using the Estimation of STromal and Immune cells in MAlignant Tumours using Expression data (ESTIMATE) algorithm. Differentially expressed lncRNAs (DElncRNAs) between low- and high-hypoxia groups, low- and high-immune groups, and tumor and control samples were identified using the limma package. Hypoxia-immune–related lncRNAs were obtained by intersecting these DElncRNAs. A hypoxia-immune–related lncRNA risk signature was developed using univariate Cox regression and least absolute shrinkage and selection operator (LASSO) analyses. The tumor microenvironments in the low- and high-risk groups were evaluated using ssGSEA, ESTIMATE, and the expression of immune checkpoints. The therapeutic response in the two groups was assessed using TIDE, IPS, and IC50. A ceRNA network based on signature lncRNAs was constructed. Finally, we used RT-qPCR to verify the expression of hypoxia-immune–related lncRNA signatures in normal and cancer tissues.ResultsUsing differential expression analysis, and univariate Cox and LASSO regression analyses, ZNF667-AS1, LINC01354, LINC00996, DANCR, CECR7, and LINC01116 were selected to construct a hypoxia-immune–related lncRNA signature. The performance of the risk signature in predicting CRC prognosis was validated in internal and external datasets, as evidenced by receiver operating characteristic curves. In addition, we observed significant differences in the tumor microenvironment and immunotherapy response between low- and high-risk groups and constructed a CECR7–miRNA–mRNA regulatory network in CRC. Furthermore, RT-qPCR results confirmed that the expression patterns of the six lncRNA signatures were consistent with those in TCGA-CRC cohort.ConclusionOur study identified six hypoxia-immune–related lncRNAs for predicting CRC survival and sensitivity to immunotherapy. These findings may enrich our understanding of CRC and help improve CRC treatment. However, large-scale long-term follow-up studies are required for verification.

show abstract

“…Traditional compound efficacy prediction models are usually based on a single type of compounds or a single target, which is difficult to use for large-scale compounddisease relationship prediction. When new compounds or new indications appear, traditional compound efficacy prediction models will be unsatisfactory [26]. In this study, based on the GoogLeNet algorithm, a CDRs prediction model was constructed for more than 4000 compounds in the CTD database and the Drugbank database.…”

Section: Discussionmentioning

confidence: 99%

“…GoogLeNet solves this problem very well. e modular structure of inception makes it increase the depth and width of the network without increasing the parameters [26]. erefore, it can achieve better prediction performance with fewer parameters.…”

Section: Introductionmentioning

confidence: 99%

A Discovery Strategy for Active Compounds of Chinese Medicine Based on the Prediction Model of Compound-Disease Relationship

Huo

Peng

et al. 2022

Journal of Oncology

View full text Add to dashboard Cite

An accurate characterization of diseases and compounds is the key to predicting the compound-disease relationship (CDR). However, due to the difficulty of a comprehensive description of CDR, the accuracy of traditional drug development models for large-scale CDR prediction is usually unsatisfactory. In order to solve this problem, we propose a new method that integrates the molecular descriptors of compounds and the symptom descriptors of diseases to build a CDR two-dimensional matrix to predict candidate active compounds. The Matlab software draws grayscale images of CDRs, which are used as a benchmark dataset for training convolutional neural network (CNN) models. The trained model is used to predict candidate antitumor active compounds. Among the AlexNet and GoogLeNet models, we selected the GoogLeNet model for the prediction of active compounds in Chinese medicine, and its Acc, Sen, Pre, F-measure, MCC, and AUC are 0.960, 0.956, 0.965, 0.960, 0.920, and 0.964, respectively. In the prediction results of compounds, 1624 candidate CDRs were found in 124 Chinese medicines. Among them, we obtained 31 features of candidate antitumor active compounds. This method provides new insights for the discovery of candidate active compounds in Chinese medicine.

show abstract

Performance Comparisons of AlexNet and GoogLeNet in Cell Growth Inhibition IC50 Prediction

Cited by 13 publications

References 58 publications

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

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

Development of a novel hypoxia-immune–related LncRNA risk signature for predicting the prognosis and immunotherapy response of colorectal cancer

A Discovery Strategy for Active Compounds of Chinese Medicine Based on the Prediction Model of Compound-Disease Relationship

Contact Info

Product

Resources

About