Virtual screening by a new Clustering-based Weighted Similarity Extreme Learning Machine approach

Pasupa, Kitsuchart; Kudisthalert, Wasu

doi:10.1371/journal.pone.0195478

Cited by 10 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are some evidences that showed the potential of using WELM classification method on one of the most challenging datasets, Maximum Unbiased Validation Dataset, which is dramatic imbalance between classes and highly diverse in structure. The experiment result showed that WELM give better performance than other standard methods, i.e., SVM, Random Forest, and Similarity Searching [38]- [40]. In this study, we demonstrated a list of contribution as follow:…”

Section: Related Workmentioning

confidence: 80%

Counting and Classification of Malarial Parasite From Giemsa-Stained Thin Film Images

2020

Self Cite

View full text Add to dashboard Cite

Malaria is a life-threatening disease causing by an infection of the protozoan parasite Plasmodium. Plasmodium falciparum is the deadliest and most common human infected parasites hosted by anopheles mosquito vector. To cure a malaria infected patient and prevent further spreading, malaria diagnosis using microscopy to visualize Giemsa-stained parasites is commonly done. The microscopy diagnosis is somewhat time consuming and requires well-trained malaria experts to interpret what they see under the microscope. To address this limitation, an automated malaria infected diagnosis is needed. This work proposed a computer-aided automated diagnosis system that can perform remote field diagnosis with high accuracy while requiring less computational demands. The proposed framework consists of two main parts that are red blood cell counting and parasite life-cycle stage classification. The counting process is performed by computer vision techniques, namely Hough transform. Different machine learning techniques, i.e., Multilayer Perceptron, Linear Discriminant Analysis, Support Vector Machine, and Weighted Similarity Extreme Learning Machine, are employed in the classification task. We also demonstrated that combining hand-crafted and deep-learned features can enhance the overall performance of the framework. The experimental results showed that the proposed methods could correctly count and classify at 97.94% and 98.12% accuracy, respectively. The overall proposal system can achieve at 96.18% accuracy. This is achieved by WELM in conjunction with deep-learned (AlexNet_FC7) and the hand-crafted (color) features. INDEX TERMS Combining features, Giemsa-stained thin film, malaria.

show abstract

Section: Related Workmentioning

confidence: 80%

Counting and Classification of Malarial Parasite From Giemsa-Stained Thin Film Images

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Specifically, to avoid time-consuming and high computationally intensive calculations, sets of templates of actives were preselected through PAM clustering analysis. The clustering algorithm is exploited in VS as a data analysis tool for uncovering representative patterns in large dataset on the basis of morphological and energetic properties. , In this work, the PAM algorithm was run to find an optimal number of consistent groups of actives: this strategy was aimed at selecting a small number of templates of actives which were labeled as medoids to be effectively employed by the LDA algorithm for classification purposes.…”

Section: Resultsmentioning

confidence: 99%

An Integrated Machine Learning Model To Spot Peptide Binding Pockets in 3D Protein Screening

Trisciuzzi

Siragusa

Baroni

et al. 2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

The prediction of peptide−protein binding sites is of utmost importance to tackle the onset of severe neurodegenerative diseases and cancer. In this work, we detail a novel machine learning model based on Linear Discriminant Analysis (LDA) demonstrating to be highly predictive in detecting the putative protein binding regions of small peptides. Starting from 439 high-quality pockets derived from peptide−protein crystallographic complexes, three sets of well-established peptidebinding regions were first selected through a Partitioning Around Medoids (PAM) clustering algorithm based on morphological and energetic 3D GRID-MIF molecular descriptors. Next, the best combination between all the putative interacting peptide pockets and related GRID-MIF scores was automatically explored by using the LDA-based protocol implemented in BioGPS. This approach proved successful to recognize the actual interacting peptide regions (that is, AUC = 0.86 and partial ROC enrichment at 5% of 0.48) from all the other pockets of the protein. Validated on two external collections sets, including 445 and 347 crystallographic peptide−protein complexes, our LDA-based model could be effective to further run peptide−protein virtual screening campaigns.

show abstract

“…WELM can reduce training time because it does not need any tuning of the kernel parameters. It yields better performance, especially when dealing with similarity-based tasks [ 32 , 41 ]. In WELM, the matrix in conventional activation is replaced by: The set of weights are randomly selected from a training set , thus, .…”

Section: Methodsmentioning

confidence: 99%

SELM: Siamese extreme learning machine with application to face biometrics

Kudisthalert

Pasupa

Morales

et al. 2022

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

Extreme learning machine (ELM) is a powerful classification method and is very competitive among existing classification methods. It is speedy at training. Nevertheless, it cannot perform face verification tasks properly because face verification tasks require the comparison of facial images of two individuals simultaneously and decide whether the two faces identify the same person. The ELM structure was not designed to feed two input data streams simultaneously. Thus, in 2-input scenarios, ELM methods are typically applied using concatenated inputs. However, this setup consumes two times more computational resources, and it is not optimized for recognition tasks where learning a separable distance metric is critical. For these reasons, we propose and develop a Siamese extreme learning machine (SELM). SELM was designed to be fed with two data streams in parallel simultaneously. It utilizes a dual-stream Siamese condition in the extra Siamese layer to transform the data before passing it to the hidden layer. Moreover, we propose a Gender-Ethnicity-dependent triplet feature exclusively trained on various specific demographic groups. This feature enables learning and extracting useful facial features of each group. Experiments were conducted to evaluate and compare the performances of SELM, ELM, and deep convolutional neural network (DCNN). The experimental results showed that the proposed feature could perform correct classification at accuracy and area under the curve (AUC). They also showed that using SELM in conjunction with the proposed feature provided accuracy and AUC. SELM outperformed the robust performances over the well-known DCNN and ELM methods.

show abstract

Virtual screening by a new Clustering-based Weighted Similarity Extreme Learning Machine approach

Cited by 10 publications

References 27 publications

Counting and Classification of Malarial Parasite From Giemsa-Stained Thin Film Images

Counting and Classification of Malarial Parasite From Giemsa-Stained Thin Film Images

An Integrated Machine Learning Model To Spot Peptide Binding Pockets in 3D Protein Screening

SELM: Siamese extreme learning machine with application to face biometrics

Contact Info

Product

Resources

About