maGENEgerZ: An Efficient AI-Based Framework Can Extract More Expressed Genes and Biological Insights Underlying Breast Cancer Drug Response Mechanism DOI Open Access
Turki Turki, Y‐h. Taguchi

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Dec. 30, 2023

Abstract Understanding breast cancer drug response mechanism can play a crucial role in improving the treatment outcomes and survival rates. Existing bioinformatics-based approaches are far from perfect do not adopt computational methods based on advanced artificial intelligence concepts. Therefore, we introduce novel framework an efficient support vector machines (esvm) working as follows. First, downloaded processed three gene expression datasets related to responding non-responding treatments omnibus (GEO) according following GEO accession numbers: GSE130787, GSE140494, GSE196093. Our method esvm is formulated constrained optimization problem dual form function of λ. We recover importance each λ, y, x. Then, select p genes out n, provided input enrichment analysis tools, Enrichr Metascape. Compared existing baseline including deep learning, results demonstrate superiority efficiency achieving high performance having more expressed well-established cell lines MD-MB231, MCF7, HS578T. Moreover, able identify (1) various drugs clinically approved ones (e.g., tamoxifen erlotinib); (2) seventy-four unique (including tumor suppression such TP53 BRCA1); (3) thirty-six TFs SP1 RELA). These have been reported be linked mechanism, progression, metastasizing. available publicly maGENEgerZ web server at https://aibio.shinyapps.io/maGENEgerZ/ .

Language: Английский

Graph convolutional network for structural equivalent key nodes identification in complex networks DOI

Asmita Patel,

Buddha Singh

Chaos Solitons & Fractals, Journal Year: 2025, Volume and Issue: 196, P. 116376 - 116376

Published: April 4, 2025

Language: Английский

Citations

0
GENEvaRX: A novel AI-driven method and web tool can identify critical genes and effective drugs for Lichen Planus DOI Open Access
Turki Turki, Y‐h. Taguchi

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 124, P. 106607 - 106607

Published: July 4, 2023

Language: Английский

Citations

3
Identification of hub genes and potential molecular mechanisms related to drug sensitivity in acute myeloid leukemia based on machine learning DOI Creative Commons
Boyu Zhang, Haiyan Liu,

Fengxia Wu

et al.

Frontiers in Pharmacology, Journal Year: 2024, Volume and Issue: 15

Published: April 8, 2024

Background: Acute myeloid leukemia (AML) is the most common form of among adults and characterized by uncontrolled proliferation clonal expansion hematopoietic cells. There has been a significant improvement in treatment younger patients, however, prognosis elderly AML patients remains poor. Methods: We used computational methods machine learning (ML) techniques to identify explore differential high-risk genes (DHRGs) AML. The DHRGs were explored through multiple silico approaches including genomic functional analysis, survival immune infiltration, miRNA co-expression stemness features analyses reveal their prognostic importance Furthermore, using different ML algorithms, models constructed validated DHRGs. At end molecular docking studies performed potential drug candidates targeting selected Results: identified total 80 comparing differentially expressed derived between normal controls Cox regression. Genetic epigenetic alteration revealed association copy number variations methylation status with overall (OS) patients. Out 137 combination Ridge plsRcox maintained highest mean C-index was build final model. When classified into low- groups based on DHRGs, low-risk group had significantly longer OS training validation cohorts. coexpression, feature hallmark pathway differences groups. Drug sensitivity top 5 drugs, carboplatin austocystin-D that may affect Conclusion: findings from current study set be as therapeutic markers for In addition, use algorithms constructing validating demonstrated. Although our extensive bioinformatics hub AML, experimental validations knock-out/-in would strengthen findings.

Language: Английский

Citations

0
maGENEgerZ: An Efficient Artificial Intelligence-Based Framework Can Extract More Expressed Genes and Biological Insights Underlying Breast Cancer Drug Response Mechanism DOI Creative Commons
Turki Turki, Y‐h. Taguchi

Mathematics, Journal Year: 2024, Volume and Issue: 12(10), P. 1536 - 1536

Published: May 15, 2024

Understanding breast cancer drug response mechanisms can play a crucial role in improving treatment outcomes and survival rates. Existing bioinformatics-based approaches are far from perfect do not adopt computational methods based on advanced artificial intelligence concepts. Therefore, we introduce novel framework an efficient support vector machine (esvm) working as follows: First, downloaded processed three gene expression datasets related to responding non-responding treatments the omnibus (GEO) according following GEO accession numbers: GSE130787, GSE140494, GSE196093. Our method esvm is formulated constrained optimization problem its dual form function of λ. We recover importance each λ, y, x. Then, select p genes out n, which provided input enrichment analysis tools, Enrichr Metascape. Compared existing baseline methods, including deep learning, results demonstrate superiority efficiency esvm, achieving high-performance having more expressed well-established cell lines, MD-MB231, MCF7, HS578T. Moreover, able identify (1) various drugs, clinically approved ones (e.g., tamoxifen erlotinib); (2) seventy-four unique (including tumor suppression such TP53 BRCA1); (3) thirty-six TFs SP1 RELA). These have been reported be linked mechanisms, progression, metastasizing. available publicly maGENEgerZ web server.

Language: Английский

Citations

0
GENEvaRX: A Novel AI-Driven Method and Web Tool Can Identify Critical Genes and Effective Drugs for Lichen Planus DOI Open Access
Turki Turki, Y‐h. Taguchi

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Feb. 24, 2023

Abstract Lichen planus (LP) is an autoimmune disorder diagnosed based on physical symptoms and lab tests. Examples of include flat bumps, itchy purplish skin, while tests a shave biopsy the lesion. When pathology report shows consistency with LP negative for potential triggers allergy test hepatitis C, dermatologist typically prescribes corticosteroid in form pills or injection into lesion to treat symptoms. To understand molecular mechanism disease thereby overcome issues associated treatment, there need identify effective drugs, drug targets, therapeutic targets LP. Hence, we propose novel computational framework new constrained optimization support vector machines coupled enrichment analysis. First, downloaded three gene expression datasets (GSE63741, GSE193351, GSE52130) pertaining healthy patients from omnibus (GEO) database. We then processed each dataset entered it our select important genes. Finally, performed analysis selected genes, reporting following results. Our methods outperformed baseline terms identifying skin tissue. Moreover, 5 drugs (including, dexamethasone, retinoic acid, quercetin), 45 unique genes (including PSMB8, KRT31, KRT16, KRT19, KRT17, COL3A1, LCE2D, LCE2A), 23 TFs NFKB1, STAT1, STAT3) reportedly related pathogenesis, treatments, targets. are publicly available GENEvaRX web server at https://aibio.shinyapps.io/GENEvaRX/ .

Language: Английский

Citations

0
An Investigation of Complex Interactions Between Genetically Determined Protein Expression and the Metabolic Phenotype of Human Islet Cells Using Deep Learning DOI
K. Padmaja,

Mukhopadhyay Debarka

SN Computer Science, Journal Year: 2023, Volume and Issue: 4(6)

Published: Oct. 6, 2023

Language: Английский

Citations

0
maGENEgerZ: An Efficient AI-Based Framework Can Extract More Expressed Genes and Biological Insights Underlying Breast Cancer Drug Response Mechanism DOI Open Access
Turki Turki, Y‐h. Taguchi

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Dec. 30, 2023

Abstract Understanding breast cancer drug response mechanism can play a crucial role in improving the treatment outcomes and survival rates. Existing bioinformatics-based approaches are far from perfect do not adopt computational methods based on advanced artificial intelligence concepts. Therefore, we introduce novel framework an efficient support vector machines (esvm) working as follows. First, downloaded processed three gene expression datasets related to responding non-responding treatments omnibus (GEO) according following GEO accession numbers: GSE130787, GSE140494, GSE196093. Our method esvm is formulated constrained optimization problem dual form function of λ. We recover importance each λ, y, x. Then, select p genes out n, provided input enrichment analysis tools, Enrichr Metascape. Compared existing baseline including deep learning, results demonstrate superiority efficiency achieving high performance having more expressed well-established cell lines MD-MB231, MCF7, HS578T. Moreover, able identify (1) various drugs clinically approved ones (e.g., tamoxifen erlotinib); (2) seventy-four unique (including tumor suppression such TP53 BRCA1); (3) thirty-six TFs SP1 RELA). These have been reported be linked mechanism, progression, metastasizing. available publicly maGENEgerZ web server at https://aibio.shinyapps.io/maGENEgerZ/ .

Language: Английский

Citations

0