High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer
The advancement of bioinformatics and machine learning has facilitated the discovery and validation of omics-based biomarkers. This study employed a novel approach combining multi-platform transcriptomics and cutting-edge algorithms to introduce novel signatures for accurate diagnosis of colorectal...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-01-01
|
| Series: | International Journal of Molecular Sciences |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1422-0067/20/2/296 |
| _version_ | 1811324135263961088 |
|---|---|
| author | Nguyen Phuoc Long Seongoh Park Nguyen Hoang Anh Tran Diem Nghi Sang Jun Yoon Jeong Hill Park Johan Lim Sung Won Kwon |
| author_facet | Nguyen Phuoc Long Seongoh Park Nguyen Hoang Anh Tran Diem Nghi Sang Jun Yoon Jeong Hill Park Johan Lim Sung Won Kwon |
| author_sort | Nguyen Phuoc Long |
| collection | DOAJ |
| description | The advancement of bioinformatics and machine learning has facilitated the discovery and validation of omics-based biomarkers. This study employed a novel approach combining multi-platform transcriptomics and cutting-edge algorithms to introduce novel signatures for accurate diagnosis of colorectal cancer (CRC). Different random forests (RF)-based feature selection methods including the area under the curve (AUC)-RF, Boruta, and Vita were used and the diagnostic performance of the proposed biosignatures was benchmarked using RF, logistic regression, naïve Bayes, and k-nearest neighbors models. All models showed satisfactory performance in which RF appeared to be the best. For instance, regarding the RF model, the following were observed: mean accuracy 0.998 (standard deviation (SD) < 0.003), mean specificity 0.999 (SD < 0.003), and mean sensitivity 0.998 (SD < 0.004). Moreover, proposed biomarker signatures were highly associated with multifaceted hallmarks in cancer. Some biomarkers were found to be enriched in epithelial cell signaling in Helicobacter pylori infection and inflammatory processes. The overexpression of TGFBI and S100A2 was associated with poor disease-free survival while the down-regulation of NR5A2, SLC4A4, and CD177 was linked to worse overall survival of the patients. In conclusion, novel transcriptome signatures to improve the diagnostic accuracy in CRC are introduced for further validations in various clinical settings. |
| first_indexed | 2024-04-13T14:09:00Z |
| format | Article |
| id | doaj.art-863f845f379a4afe9c135acf513cecdf |
| institution | Directory Open Access Journal |
| issn | 1422-0067 |
| language | English |
| last_indexed | 2024-04-13T14:09:00Z |
| publishDate | 2019-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | International Journal of Molecular Sciences |
| spelling | doaj.art-863f845f379a4afe9c135acf513cecdf2022-12-22T02:43:50ZengMDPI AGInternational Journal of Molecular Sciences1422-00672019-01-0120229610.3390/ijms20020296ijms20020296High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal CancerNguyen Phuoc Long0Seongoh Park1Nguyen Hoang Anh2Tran Diem Nghi3Sang Jun Yoon4Jeong Hill Park5Johan Lim6Sung Won Kwon7College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 08826, KoreaDepartment of Statistics, Seoul National University, Seoul 08826, KoreaCollege of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 08826, KoreaSchool of Medicine, Vietnam National University, Ho Chi Minh 70000, VietnamCollege of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 08826, KoreaCollege of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 08826, KoreaDepartment of Statistics, Seoul National University, Seoul 08826, KoreaCollege of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 08826, KoreaThe advancement of bioinformatics and machine learning has facilitated the discovery and validation of omics-based biomarkers. This study employed a novel approach combining multi-platform transcriptomics and cutting-edge algorithms to introduce novel signatures for accurate diagnosis of colorectal cancer (CRC). Different random forests (RF)-based feature selection methods including the area under the curve (AUC)-RF, Boruta, and Vita were used and the diagnostic performance of the proposed biosignatures was benchmarked using RF, logistic regression, naïve Bayes, and k-nearest neighbors models. All models showed satisfactory performance in which RF appeared to be the best. For instance, regarding the RF model, the following were observed: mean accuracy 0.998 (standard deviation (SD) < 0.003), mean specificity 0.999 (SD < 0.003), and mean sensitivity 0.998 (SD < 0.004). Moreover, proposed biomarker signatures were highly associated with multifaceted hallmarks in cancer. Some biomarkers were found to be enriched in epithelial cell signaling in Helicobacter pylori infection and inflammatory processes. The overexpression of TGFBI and S100A2 was associated with poor disease-free survival while the down-regulation of NR5A2, SLC4A4, and CD177 was linked to worse overall survival of the patients. In conclusion, novel transcriptome signatures to improve the diagnostic accuracy in CRC are introduced for further validations in various clinical settings.http://www.mdpi.com/1422-0067/20/2/296colorectal cancertranscriptomicsdiagnosisbiomarkermachine learningvariable selection |
| spellingShingle | Nguyen Phuoc Long Seongoh Park Nguyen Hoang Anh Tran Diem Nghi Sang Jun Yoon Jeong Hill Park Johan Lim Sung Won Kwon High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer International Journal of Molecular Sciences colorectal cancer transcriptomics diagnosis biomarker machine learning variable selection |
| title | High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer |
| title_full | High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer |
| title_fullStr | High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer |
| title_full_unstemmed | High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer |
| title_short | High-Throughput Omics and Statistical Learning Integration for the Discovery and Validation of Novel Diagnostic Signatures in Colorectal Cancer |
| title_sort | high throughput omics and statistical learning integration for the discovery and validation of novel diagnostic signatures in colorectal cancer |
| topic | colorectal cancer transcriptomics diagnosis biomarker machine learning variable selection |
| url | http://www.mdpi.com/1422-0067/20/2/296 |
| work_keys_str_mv | AT nguyenphuoclong highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT seongohpark highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT nguyenhoanganh highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT trandiemnghi highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT sangjunyoon highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT jeonghillpark highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT johanlim highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer AT sungwonkwon highthroughputomicsandstatisticallearningintegrationforthediscoveryandvalidationofnoveldiagnosticsignaturesincolorectalcancer |