1
|
Scherr TF, Douglas CE, Schaecher KE, Schoepp RJ, Ricks KM, Shoemaker CJ. Application of a Machine Learning-Based Classification Approach for Developing Host Protein Diagnostic Models for Infectious Disease. Diagnostics (Basel) 2024; 14:1290. [PMID: 38928705 PMCID: PMC11202442 DOI: 10.3390/diagnostics14121290] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/09/2024] [Revised: 06/11/2024] [Accepted: 06/13/2024] [Indexed: 06/28/2024] Open
Abstract
In recent years, infectious disease diagnosis has increasingly turned to host-centered approaches as a complement to pathogen-directed ones. The former, however, typically requires the interpretation of complex multiple biomarker datasets to arrive at an informative diagnostic outcome. This report describes a machine learning (ML)-based classification workflow that is intended as a template for researchers seeking to apply ML approaches for developing host-based infectious disease biomarker classifiers. As an example, we built a classification model that could accurately distinguish between three disease etiology classes: bacterial, viral, and normal in human sera using host protein biomarkers of known diagnostic utility. After collecting protein data from known disease samples, we trained a series of increasingly complex Auto-ML models until arriving at an optimized classifier that could differentiate viral, bacterial, and non-disease samples. Even when limited to a relatively small training set size, the model had robust diagnostic characteristics and performed well when faced with a blinded sample set. We present here a flexible approach for applying an Auto-ML-based workflow for the identification of host biomarker classifiers with diagnostic utility for infectious disease, and which can readily be adapted for multiple biomarker classes and disease states.
Collapse
Affiliation(s)
| | - Christina E. Douglas
- Diagnostic Systems Division, U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD 21702, USA (R.J.S.); (K.M.R.)
| | - Kurt E. Schaecher
- Virology Division, U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD 21702, USA
| | - Randal J. Schoepp
- Diagnostic Systems Division, U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD 21702, USA (R.J.S.); (K.M.R.)
| | - Keersten M. Ricks
- Diagnostic Systems Division, U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD 21702, USA (R.J.S.); (K.M.R.)
| | - Charles J. Shoemaker
- Diagnostic Systems Division, U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD 21702, USA (R.J.S.); (K.M.R.)
| |
Collapse
|
2
|
Bostanci E, Kocak E, Unal M, Guzel MS, Acici K, Asuroglu T. Machine Learning Analysis of RNA-seq Data for Diagnostic and Prognostic Prediction of Colon Cancer. SENSORS (BASEL, SWITZERLAND) 2023; 23:3080. [PMID: 36991790 PMCID: PMC10052105 DOI: 10.3390/s23063080] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/09/2023] [Revised: 03/09/2023] [Accepted: 03/11/2023] [Indexed: 06/19/2023]
Abstract
Data from omics studies have been used for prediction and classification of various diseases in biomedical and bioinformatics research. In recent years, Machine Learning (ML) algorithms have been used in many different fields related to healthcare systems, especially for disease prediction and classification tasks. Integration of molecular omics data with ML algorithms has offered a great opportunity to evaluate clinical data. RNA sequence (RNA-seq) analysis has been emerged as the gold standard for transcriptomics analysis. Currently, it is being used widely in clinical research. In our present work, RNA-seq data of extracellular vesicles (EV) from healthy and colon cancer patients are analyzed. Our aim is to develop models for prediction and classification of colon cancer stages. Five different canonical ML and Deep Learning (DL) classifiers are used to predict colon cancer of an individual with processed RNA-seq data. The classes of data are formed on the basis of both colon cancer stages and cancer presence (healthy or cancer). The canonical ML classifiers, which are k-Nearest Neighbor (kNN), Logistic Model Tree (LMT), Random Tree (RT), Random Committee (RC), and Random Forest (RF), are tested with both forms of the data. In addition, to compare the performance with canonical ML models, One-Dimensional Convolutional Neural Network (1-D CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) DL models are utilized. Hyper-parameter optimizations of DL models are constructed by using genetic meta-heuristic optimization algorithm (GA). The best accuracy in cancer prediction is obtained with RC, LMT, and RF canonical ML algorithms as 97.33%. However, RT and kNN show 95.33% performance. The best accuracy in cancer stage classification is achieved with RF as 97.33%. This result is followed by LMT, RC, kNN, and RT with 96.33%, 96%, 94.66%, and 94%, respectively. According to the results of the experiments with DL algorithms, the best accuracy in cancer prediction is obtained with 1-D CNN as 97.67%. BiLSTM and LSTM show 94.33% and 93.67% performance, respectively. In classification of the cancer stages, the best accuracy is achieved with BiLSTM as 98%. 1-D CNN and LSTM show 97% and 94.33% performance, respectively. The results reveal that both canonical ML and DL models may outperform each other for different numbers of features.
Collapse
Affiliation(s)
- Erkan Bostanci
- Department of Computer Engineering, Faculty of Engineering, Ankara University, 06830 Ankara, Turkey
| | - Engin Kocak
- Department of Analytical Chemistry, Faculty of Gülhane Pharmacy, University of Health Sciences, 06018 Ankara, Turkey
| | - Metehan Unal
- Department of Computer Engineering, Faculty of Engineering, Ankara University, 06830 Ankara, Turkey
| | - Mehmet Serdar Guzel
- Department of Computer Engineering, Faculty of Engineering, Ankara University, 06830 Ankara, Turkey
| | - Koray Acici
- Department of Artificial Intelligence and Data Engineering, Faculty of Engineering, Ankara University, 06830 Ankara, Turkey
| | - Tunc Asuroglu
- Faculty of Medicine and Health Technology, Tampere University, 33720 Tampere, Finland
| |
Collapse
|
3
|
ALAKUŞ TB, TÜRKOĞLU İ. Kanser Teşhisinde Protein Haritalama Tekniklerinin Başarımlarının Derin Öğrenme Kullanılarak Karşılaştırılması. FIRAT ÜNIVERSITESI MÜHENDISLIK BILIMLERI DERGISI 2021; 33:547-565. [DOI: 10.35234/fumbd.881228] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 09/01/2023]
Abstract
Kanser, dünya çapında çoğu insanın ölmesine neden olan ve birçok farklı alt tiplerden oluşan heterojen bir hastalıktır. Bir kanser türünün erken teşhisi ve prognozu, hastaların sonraki klinik takibini kolaylaştırabildiği için kanser araştırmalarında bir gereklilik haline gelmiştir. Bunun için en çok kullanılan yöntemlerden birisi histolojik incelemedir. Ancak bu yöntemde çok sayıda gözlemciler arası değişkenlik bulunmakta, bu ise inceleme sürecinin uzun olmasına ve zaman almasına neden olmaktadır. Bu dezavantajın önüne geçmek için araştırmacılar hesaplama-tabanlı yaklaşımlara yönelmişler ve kanserli proteinlerin belirlenmesi için protein-protein etkileşimleri, protein etkileşim ağları ve moleküler parmak izleri yöntemlerinden yararlanmaktadırlar. Bu yöntemler arasında, çeşitli çalışmalar genomik bilgilerden de kanserli hücrelerin tespit edilebildiğini göstermiştir. Kansere ait genlerin dizilimlerine göre belirli kanser türlerinin belirlenebildiği ve bu süreçte yapay öğrenme tabanlı yaklaşımların etkili olduğu görülmüştür. Bu çalışmada, derin öğrenme algoritmalarından birisi olan tekrarlayıcı sinir ağı mimarisi kullanılmış ve insana ait mesane, kolon ve prostat kanserlerinin, protein dizilimlerine göre sınıflandırılması yapılmıştır. Çalışma, verilerin elde edilmesi, protein dizilimlerinin sayısallaştırılması, derin öğrenme model uygulamasının geliştirilmesi ve protein haritalama tekniklerinin başarımının karşılaştırılması olmak üzere dört aşamadan meydana gelmektedir. Protein dizilimlerini sayısallaştırmak için AESNN1, hidrofobiklik, tam sayı, Miyazawa enerjileri ve rastgele kodlama yöntemleri ele alınmıştır. Çalışmanın sonunda, mesane kanseri için en yüksek doğruluk değeri %87.15 ile AESNN1 haritalama yöntemiyle, kolon kanseri ve prostat kanseri için ise en yüksek doğruluk değeri sırasıyla %94.40 ve %75.45 olarak Miyazawa enerjileri ve rastgele kodlama protein haritalama yöntemi ile elde edilmiştir. Bu çalışma ile yapay öğrenme ve protein haritalama tekniklerinin, kanserli protein dizilimlerinin belirlenmesinde etkili olduğu gözlemlenmiştir.
Collapse
|
4
|
K T, N KV, S S. Distribution based Fuzzy Estimate Spectral Clustering for Cancer Detection with Protein Sequence and Structural Motifs. Asian Pac J Cancer Prev 2018; 19:1935-1940. [PMID: 30051675 PMCID: PMC6165630 DOI: 10.22034/apjcp.2018.19.7.1935] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Abstract] [Key Words] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/14/2022] Open
Abstract
Objective: In biological data analysis, protein sequence and structural motifs are an amino-acid sequence patterns
that are widespread and used as tools for detecting the cancer at an earlier stage. To improve the cancer detection with
minimum space and time complexity, Distribution based Fuzzy Estimate Spectral Clustering (DFESC) technique is
developed. Methods: Initially, the protein sequence motifs are taken from dataset to form the cluster. The Distribution
based spectral clustering is applied to group the protein sequence by measuring the generalized jaccard similarity
between each protein sequences. To develop the clustering accuracy, soft computing technique namely fuzzy logic is
applied to calculate membership value of each sequence motifs. Results: The outcome showed that the presented DFESC
technique effectively identifies the cancer in terms of clustering accuracy, false positive rate, and cancer detection time
and space complexity. Conclusion: Based on the observations, evaluation of DFESC technique provides improved
result for premature detection of cancer using protein sequence and structural motifs.
Collapse
Affiliation(s)
- Thenmozhi K
- Department of Computer Applications, Selvam College of Technology, Namakkal, TamilNadu, India,For Correspondence:
| | | | - Shanthi S
- Department of Computer Applications, Kongu Engineering College, Erode, TamilNadu, India
| |
Collapse
|
5
|
Vyas R, Bapat S, Goel P, Karthikeyan M, Tambe SS, Kulkarni BD. Application of Genetic Programming (GP) Formalism for Building Disease Predictive Models from Protein-Protein Interactions (PPI) Data. IEEE/ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2018; 15:27-37. [PMID: 28113781 DOI: 10.1109/tcbb.2016.2621042] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/06/2023]
Abstract
Protein-protein interactions (PPIs) play a vital role in the biological processes involved in the cell functions and disease pathways. The experimental methods known to predict PPIs require tremendous efforts and the results are often hindered by the presence of a large number of false positives. Herein, we demonstrate the use of a new Genetic Programming (GP) based Symbolic Regression (SR) approach for predicting PPIs related to a disease. In a case study, a dataset consisting of one hundred and thirty five PPI complexes related to cancer was used to construct a generic PPI predicting model with good PPI prediction accuracy and generalization ability. A high correlation coefficient(CC) of 0.893, low root mean square error (RMSE) and mean absolute percentage error (MAPE) values of 478.221 and 0.239, respectively were achieved for both the training and test set outputs. To validate the discriminatory nature of the model, it was applied on a dataset of diabetes complexes where it yielded significantly low CC values. Thus, the GP model developed here serves a dual purpose: (a)a predictor of the binding energy of cancer related PPI complexes, and (b)a classifier for discriminating PPI complexes related to cancer from those of other diseases.
Collapse
|
6
|
Zhou B, Sun Q, Kong DX. Predicting cancer-relevant proteins using an improved molecular similarity ensemble approach. Oncotarget 2017; 7:32394-407. [PMID: 27083051 PMCID: PMC5078021 DOI: 10.18632/oncotarget.8716] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/26/2015] [Accepted: 03/28/2016] [Indexed: 12/15/2022] Open
Abstract
In this study, we proposed an improved algorithm for identifying proteins relevant to cancer. The algorithm was named two-layer molecular similarity ensemble approach (TL-SEA). We applied TL-SEA to analyzing the correlation between anticancer compounds (against cell lines K562, MCF7 and A549) and active compounds against separate target proteins listed in BindingDB. Several associations between cancer types and related proteins were revealed using this chemoinformatics approach. An analysis of the literature showed that 26 of 35 predicted proteins were correlated with cancer cell proliferation, apoptosis or differentiation. Additionally, interactions between proteins in BindingDB and anticancer chemicals were also predicted. We discuss the roles of the most important predicted proteins in cancer biology and conclude that TL-SEA could be a useful tool for inferring novel proteins involved in cancer and revealing underlying molecular mechanisms.
Collapse
Affiliation(s)
- Bin Zhou
- State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan 430070, China.,Agricultural Bioinformatics Key Laboratory of Hubei Province, College of Informatics, Huazhong Agricultural University, Wuhan 430070, China
| | - Qi Sun
- State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan 430070, China.,Agricultural Bioinformatics Key Laboratory of Hubei Province, College of Informatics, Huazhong Agricultural University, Wuhan 430070, China
| | - De-Xin Kong
- State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan 430070, China.,Agricultural Bioinformatics Key Laboratory of Hubei Province, College of Informatics, Huazhong Agricultural University, Wuhan 430070, China
| |
Collapse
|
7
|
Huang CH, Chang PMH, Hsu CW, Huang CYF, Ng KL. Drug repositioning for non-small cell lung cancer by using machine learning algorithms and topological graph theory. BMC Bioinformatics 2016; 17 Suppl 1:2. [PMID: 26817825 PMCID: PMC4895785 DOI: 10.1186/s12859-015-0845-0] [Citation(s) in RCA: 15] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/12/2022] Open
Abstract
Background Non-small cell lung cancer (NSCLC) is one of the leading causes of death globally, and research into NSCLC has been accumulating steadily over several years. Drug repositioning is the current trend in the pharmaceutical industry for identifying potential new uses for existing drugs and accelerating the development process of drugs, as well as reducing side effects. Results This work integrates two approaches - machine learning algorithms and topological parameter-based classification - to develop a novel pipeline of drug repositioning to analyze four lung cancer microarray datasets, enriched biological processes, potential therapeutic drugs and targeted genes for NSCLC treatments. A total of 7 (8) and 11 (12) promising drugs (targeted genes) were discovered for treating early- and late-stage NSCLC, respectively. The effectiveness of these drugs is supported by the literature, experimentally determined in-vitro IC50 and clinical trials. This work provides better drug prediction accuracy than competitive research according to IC50 measurements. Conclusions With the novel pipeline of drug repositioning, the discovery of enriched pathways and potential drugs related to NSCLC can provide insight into the key regulators of tumorigenesis and the treatment of NSCLC. Based on the verified effectiveness of the targeted drugs predicted by this pipeline, we suggest that our drug-finding pipeline is effective for repositioning drugs.
Collapse
Affiliation(s)
- Chien-Hung Huang
- Department of Computer Science and Information Engineering, National Formosa University, Hu-Wei, 63205, Taiwan.
| | - Peter Mu-Hsin Chang
- Division of Hematology and Oncology, Department of Medicine, Taipei Veterans General Hospital; Faculty of Medicine, National Yang Ming University, Taipei, 112, Taiwan.
| | - Chia-Wei Hsu
- Department of Computer Science and Information Engineering, National Formosa University, Hu-Wei, 63205, Taiwan.
| | - Chi-Ying F Huang
- Institute of Biopharmaceutical Sciences, National Yang-Ming University, Taipei, 112, Taiwan.
| | - Ka-Lok Ng
- Department of Bioinformatics and Medical Engineering, Asia University, Taichung, 41354, Taiwan. .,Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, 40402, Taiwan.
| |
Collapse
|