Taba: A Tool to Analyze the Binding Affinity

Silva, Amauri Duarte da; Bitencourt-Ferreira, Gabriela; Azevedo, Walter Filgueira de

doi:10.1002/jcc.26048

Cited by 31 publications

(25 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each dataset, the number of clusters used by the k-means algorithm in the experiment was chosen within the range of length 2 around the number of cell types in the dataset. For the above three datasets of 10X PBMC, Mouse Bladder Cells and Worm Neuron Cells, the selection range of the number of clusters fall into [6,10], [14,18] and [8,12] respectively. The clustering performance on these three datasets is shown in Fig.…”

Section: Selections Of the Number Of Clustersmentioning

confidence: 99%

“…Although clustering is a traditional machine learning research field [8]- [10] and there have been some representative methods such as k-means [11] and spectral clustering [12], clustering analysis on scRNA-seq data is still a challenge due to the dropout occurring in the raw data [13]. The dropout refers to the fact that there are some false-zero counts and gene count matrix may contain actually no reported data, which are caused by low sequencing depth and other technology limits.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

AAE-SC: A scRNA-Seq Clustering Framework Based on Adversarial Autoencoder

Guo

Xiao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Single-cell RNA sequencing (scRNA-seq) provides the expression profiles of individual cells, and it is expected to provide higher cellular differential resolution than traditional bulk RNA sequencing. In scRNA-seq analysis, clustering is crucial for identifying cell types, and can be potentially exploited to understand high-level biological processes. Recently, autoencoder has been successfully applied in scRNAseq clustering problem and achieved promising results. Most existing works focus on characterizing the sparsity of data, and directly utilize the bottleneck feature of the autoencoder for clustering might not be optimal. In this paper, a novel framework named Adversarial AutoEncoder ScRNA-seq Clustering (AAE-SC) is proposed to bring an additional constraint on the bottleneck feature. Specifically, AAE-SC adds an AAE module on top of the bottleneck layer, and constrains the bottleneck feature distribution to be aligned with a consistent distribution. Also, the AAE and the reconstructed modules are jointly optimized to generate a highly discriminative and consistent feature, which is further proceeded for clustering. We find that by using AAE-SC to impose certain constraints on the features of the hidden layer, the performance of clustering can be improved. Experimental results on three real-world datasets demonstrate that the proposed AAE-SC framework outperformed state-of-the-art methods by 2% at least and 5% at most. And AAE-SC shows more robustness than the baseline model for downsampled and unbalanced cluster size datasets. INDEX TERMS Single-cell RNA-seq data, Adversarial autoencoder, Clustering analysis, Unsupervised learning Author et al.: AAE-SC: A scRNA-seq Clustering Framework based on Adversarial Autoencoder Genes Cells Low High Cluster1 Cluster2 Cluster3 Blood Cells Bladder Cells Neurons Unsupervised Clustering Cell Label Identification FIGURE 1. Explanation of scRNA-seq clustering task. Because dropout causes the gene expression level of the original data to be very low, the data is highly sparse, and it brings difficulties to the subsequent clustering. Therefore, special clustering algorithms are required to process this type of data and correctly assign different cell samples to different And identify the cell type. The heat map on the left of the figure is a visual representation of raw scRNA-seq data, and the numbers in the heat map indicate the expression value of each gene in the cell sample. The color bars in the figure indicate the level of gene expression.

show abstract

Section: Selections Of the Number Of Clustersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

AAE-SC: A scRNA-Seq Clustering Framework Based on Adversarial Autoencoder

Guo

Xiao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Among in-silico approaches to accurately predict the effect of mutations on change in binding affinity, machine learning is preferable because of its implicit treatment of unknown factors involved in protein interaction and its ability to learn a flexible data-driven function (Ain et al, 2015;Silva et al, 2020;Xavier et al, 2016). A bunch of machine learning based methods has been proposed in the literature to predict the change in binding free energy (∆∆𝐺) upon mutations (Berliner et al, 2014;Brender and Zhang, 2015;Chen et al, 2019;Geng et al, 2019;Li et al, 2016;Pahari et al, 2020;Pires et al, 2014;Rodrigues et al, 2019;Witvliet et al, 2016;Zhang et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

PANDA: Predicting the change in proteins binding affinity upon mutations by finding a signal in primary structures

Abbasi

Abbas

Andleeb

2021

J. Bioinform. Comput. Biol.

View full text Add to dashboard Cite

Accurately determining a change in protein binding affinity upon mutations is important to find novel therapeutics and to assist mutagenesis studies. Determination of change in binding affinity upon mutations requires sophisticated, expensive, and time-consuming wet-lab experiments that can be supported with computational methods. Most of the available computational prediction techniques depend upon protein structures that bound their applicability to only protein complexes with recognized 3D structures. In this work, we explore the sequence-based prediction of change in protein binding affinity upon mutation and question the effectiveness of [Formula: see text]-fold cross-validation (CV) across mutations adopted in previous studies to assess the generalization ability of such predictors with no known mutation during training. We have used protein sequence information instead of protein structures along with machine learning techniques to accurately predict the change in protein binding affinity upon mutation. Our proposed sequence-based novel change in protein binding affinity predictor called PANDA performs comparably to the existing methods gauged through an appropriate CV scheme and an external independent test dataset. On an external test dataset, our proposed method gives a maximum Pearson correlation coefficient of 0.52 in comparison to the state-of-the-art existing protein structure-based method called MutaBind which gives a maximum Pearson correlation coefficient of 0.59. Our proposed protein sequence-based method, to predict a change in binding affinity upon mutations, has wide applicability and comparable performance in comparison to existing protein structure-based methods. We made PANDA easily accessible through a cloud-based webserver and python code available at https://sites.google.com/view/wajidarshad/software and https://github.com/wajidarshad/panda , respectively.

show abstract

“…In this study, virtual screening of the database will be done by building a virtual screening workflow [14][15][16][17][18], which uses the QSAR classification models to predict active compounds from the database and then uses the QSAR regression method to see the value of its activity [16]. The progress of QSAR methods can benefit from modern artificial intelligence (AI) approaches to develop computation models [19][20][21][22], this study focuses on the development of a QSAR method based on AI with a supervised learning algorithm. With this method, the prediction of thousands or even millions of molecules activity from a database can be made faster than other virtual screening methods [23].…”

Section: Introductionmentioning

confidence: 99%

Virtual Screening of DPP-4 Inhibitors Using QSAR-Based Artificial Intelligence and Molecular Docking of Hit Compounds to DPP-8 and DPP-9 Enzymes

Hermansyah

Bustamam

Yanuar

2020

Preprint

View full text Add to dashboard Cite

Background: Dipeptidyl Peptidase-4 (DPP-4) inhibitors are becoming an essential drug in the treatment of type 2 diabetes mellitus, but some classes of these drugs have side effects such as joint pain that can become severe to pancreatitis. It is thought that these side effects appear related to their inhibition against enzymes DPP-8 and DPP-9. Objective: This study aims to find DPP-4 inhibitor hit compounds that are selective against the DPP-8 and DPP-9 enzymes. By building a virtual screening workflow using the Quantitative Structure-Activity Relationship (QSAR) method based on artificial intelligence (AI), millions of molecules from the database can be screened for the DPP-4 enzyme target with a faster time compared to other screening methods. Result: Five regression machine learning algorithms and four classification machine learning algorithms were used to build virtual screening workflows. The algorithm that qualifies for the regression QSAR model was Support Vector regression with R 2 pred 0.78, while the classification QSAR model was Random Forest with 92.21% accuracy. The virtual screening results of more than 10 million molecules from the database, obtained 2,716 hit compounds with pIC50 above 7.5. Molecular docking results of several potential hit compounds to the DPP-4, DPP-8 and DPP-9 enzymes, obtained CH0002 hit compound that has a high inhibitory potential against the DPP-4 enzyme and low inhibition of the DPP-8 and DPP-9 enzymes. Conclusion: This research was able to produce DPP-4 inhibitor hit compounds that are potential to DPP-4 and selective to DPP-8 and DPP-9 enzymes so that they can be further developed in the DPP-4 inhibitors discovery. The resulting virtual screening workflow can be applied to the discovery of hit compounds on other targets.

show abstract

Taba: A Tool to Analyze the Binding Affinity

Cited by 31 publications

References 22 publications

AAE-SC: A scRNA-Seq Clustering Framework Based on Adversarial Autoencoder

AAE-SC: A scRNA-Seq Clustering Framework Based on Adversarial Autoencoder

PANDA: Predicting the change in proteins binding affinity upon mutations by finding a signal in primary structures

Virtual Screening of DPP-4 Inhibitors Using QSAR-Based Artificial Intelligence and Molecular Docking of Hit Compounds to DPP-8 and DPP-9 Enzymes

Contact Info

Product

Resources

About