scholarly journals Validation strategies for target prediction methods

2019 ◽  
Vol 21 (3) ◽  
pp. 791-802 ◽  
Author(s):  
Neann Mathai ◽  
Ya Chen ◽  
Johannes Kirchmair
Keyword(s):  
Small Molecules ◽  
Large Scale ◽  
Predictive Power ◽  
Molecular Similarity ◽  
Prediction Method ◽  
Target Prediction ◽  
Structural Data ◽  
Prediction Methods ◽  
Action Identification ◽  
Statistical Validation

Abstract Computational methods for target prediction, based on molecular similarity and network-based approaches, machine learning, docking and others, have evolved as valuable and powerful tools to aid the challenging task of mode of action identification for bioactive small molecules such as drugs and drug-like compounds. Critical to discerning the scope and limitations of a target prediction method is understanding how its performance was evaluated and reported. Ideally, large-scale prospective experiments are conducted to validate the performance of a model; however, this expensive and time-consuming endeavor is often not feasible. Therefore, to estimate the predictive power of a method, statistical validation based on retrospective knowledge is commonly used. There are multiple statistical validation techniques that vary in rigor. In this review we discuss the validation strategies employed, highlighting the usefulness and constraints of the validation schemes and metrics that are employed to measure and describe performance. We address the limitations of measuring only generalized performance, given that the underlying bioactivity and structural data are biased towards certain small-molecule scaffolds and target families, and suggest additional aspects of performance to consider in order to produce more detailed and realistic estimates of predictive power. Finally, we describe the validation strategies that were employed by some of the most thoroughly validated and accessible target prediction methods.

Epigenetic Target Prediction with Accurate Machine Learning Models

2021 ◽  
Author(s):  
Norberto Sánchez-Cruz ◽  
Jose L. Medina-Franco
Keyword(s):  
Machine Learning ◽  
Small Molecules ◽  
Predictive Models ◽  
Large Scale ◽  
Target Prediction ◽  
Quantitative Measure ◽  
Learning Models ◽  
Discovery Research ◽  
Drug Discovery Research ◽  
Machine Learning Models

<p>Epigenetic targets are a significant focus for drug discovery research, as demonstrated by the eight approved epigenetic drugs for treatment of cancer and the increasing availability of chemogenomic data related to epigenetics. This data represents a large amount of structure-activity relationships that has not been exploited thus far for the development of predictive models to support medicinal chemistry efforts. Herein, we report the first large-scale study of 26318 compounds with a quantitative measure of biological activity for 55 protein targets with epigenetic activity. Through a systematic comparison of machine learning models trained on molecular fingerprints of different design, we built predictive models with high accuracy for the epigenetic target profiling of small molecules. The models were thoroughly validated showing mean precisions up to 0.952 for the epigenetic target prediction task. Our results indicate that the herein reported models have considerable potential to identify small molecules with epigenetic activity. Therefore, our results were implemented as freely accessible and easy-to-use web application.</p>

scholarly journals Large-scale comparison of machine learning methods for drug target prediction on ChEMBL

2018 ◽  
Vol 9 (24) ◽  
pp. 5441-5451 ◽  
Author(s):  
Andreas Mayr ◽  
Günter Klambauer ◽  
Thomas Unterthiner ◽  
Marvin Steijaert ◽  
Jörg K. Wegner ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Comparative Study ◽  
Drug Target ◽  
Large Scale ◽  
State Of The Art ◽  
Target Prediction ◽  
Prediction Methods ◽  
Machine Learning Methods ◽  
Drug Target Prediction

The to date largest comparative study of nine state-of-the-art drug target prediction methods finds that deep learning outperforms all other competitors. The results are based on a benchmark of 1300 assays and half a million compounds.

Epigenetic Target Prediction with Accurate Machine Learning Models

2021 ◽  
Author(s):  
Norberto Sánchez-Cruz ◽  
Jose L. Medina-Franco
Keyword(s):  
Machine Learning ◽  
Small Molecules ◽  
Predictive Models ◽  
Large Scale ◽  
Target Prediction ◽  
Quantitative Measure ◽  
Learning Models ◽  
Discovery Research ◽  
Drug Discovery Research ◽  
Machine Learning Models

<p>Epigenetic targets are a significant focus for drug discovery research, as demonstrated by the eight approved epigenetic drugs for treatment of cancer and the increasing availability of chemogenomic data related to epigenetics. This data represents a large amount of structure-activity relationships that has not been exploited thus far for the development of predictive models to support medicinal chemistry efforts. Herein, we report the first large-scale study of 26318 compounds with a quantitative measure of biological activity for 55 protein targets with epigenetic activity. Through a systematic comparison of machine learning models trained on molecular fingerprints of different design, we built predictive models with high accuracy for the epigenetic target profiling of small molecules. The models were thoroughly validated showing mean precisions up to 0.952 for the epigenetic target prediction task. Our results indicate that the herein reported models have considerable potential to identify small molecules with epigenetic activity. Therefore, our results were implemented as freely accessible and easy-to-use web application.</p>

scholarly journals Analyzing the Bills-Voting Dynamics and Predicting Corruption-Convictions Among Brazilian Congressmen Through Temporal Networks

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Tiago Colliri ◽  
Liang Zhao
Keyword(s):  
Link Prediction ◽  
Large Scale ◽  
Prediction Method ◽  
High Accuracy ◽  
Financial Information ◽  
Prediction Methods ◽  
Temporal Networks ◽  
Public Data ◽  
Financial Crimes ◽  
Prediction Techniques

AbstractIn this paper, we propose a network-based technique to analyze bills-voting data comprising the votes of Brazilian congressmen for a period of 28 years. The voting sessions are initially mapped into static networks, where each node represents a congressman and each edge stands for the similarity of votes between a pair of congressmen. Afterwards, the constructed static networks are converted to temporal networks. Our analyses on the temporal networks capture some of the main political changes happened in Brazil during the period of time under consideration. Moreover, we find out that the bills-voting networks can be used to identify convicted politicians, who commit corruption or other financial crimes. Therefore, we propose two conviction prediction methods, one is based on the highest weighted convicted neighbor and the other is based on link prediction techniques. It is a surprise to us that the high accuracy (up to 90% by the link prediction method) on predicting convictions is achieved only through bills-voting data, without taking into account any financial information beforehand. Such a feature makes possible to monitor congressmen just by considering their legal public activities. In this way, our work contributes to the large scale public data study using complex networks.

scholarly journals SimilarityLab: Molecular Similarity for SAR Exploration and Target Prediction on the Web

Processes ◽  
2021 ◽  
Vol 9 (9) ◽  
pp. 1520
Author(s):  
Steven Shave ◽  
Manfred Auer
Keyword(s):  
Drug Discovery ◽  
Chemical Synthesis ◽  
Small Molecules ◽  
Similarity Measure ◽  
Molecular Similarity ◽  
Chemical Space ◽  
Source Code ◽  
Target Prediction ◽  
Structure Activity ◽  
The Web

Exploration of chemical space around hit, experimental, and known active compounds is an important step in the early stages of drug discovery. In academia, where access to chemical synthesis efforts is restricted in comparison to the pharma-industry, hits from primary screens are typically followed up through purchase and testing of similar compounds, before further funding is sought to begin medicinal chemistry efforts. Rapid exploration of druglike similars and structure–activity relationship profiles can be achieved through our new webservice SimilarityLab. In addition to searching for commercially available molecules similar to a query compound, SimilarityLab also enables the search of compounds with recorded activities, generating consensus counts of activities, which enables target and off-target prediction. In contrast to other online offerings utilizing the USRCAT similarity measure, SimilarityLab’s set of commercially available small molecules is consistently updated, currently containing over 12.7 million unique small molecules, and not relying on published databases which may be many years out of date. This ensures researchers have access to up-to-date chemistries and synthetic processes enabling greater diversity and access to a wider area of commercial chemical space. All source code is available in the SimilarityLab source repository.

scholarly journals A novel ligand of the translationally controlled tumor protein (TCTP) identified by virtual drug screening for cancer differentiation therapy

2021 ◽  
Author(s):  
Nicolas Fischer ◽  
Ean-Jeong Seo ◽  
Sara Abdelfatah ◽  
Edmond Fleischer ◽  
Anette Klinger ◽  
...  
Keyword(s):  
Cell Cycle ◽  
Virtual Screening ◽  
Small Molecules ◽  
Large Scale ◽  
Differentiation Therapy ◽  
P53 Expression ◽  
Microscale Thermophoresis ◽  
Cycle Arrest ◽  
G1 Cell Cycle Arrest ◽  
Mcf 7

SummaryIntroduction Differentiation therapy is a promising strategy for cancer treatment. The translationally controlled tumor protein (TCTP) is an encouraging target in this context. By now, this field of research is still at its infancy, which motivated us to perform a large-scale screening for the identification of novel ligands of TCTP. We studied the binding mode and the effect of TCTP blockade on the cell cycle in different cancer cell lines. Methods Based on the ZINC-database, we performed virtual screening of 2,556,750 compounds to analyze the binding of small molecules to TCTP. The in silico results were confirmed by microscale thermophoresis. The effect of the new ligand molecules was investigated on cancer cell survival, flow cytometric cell cycle analysis and protein expression by Western blotting and co-immunoprecipitation in MOLT-4, MDA-MB-231, SK-OV-3 and MCF-7 cells. Results Large-scale virtual screening by PyRx combined with molecular docking by AutoDock4 revealed five candidate compounds. By microscale thermophoresis, ZINC10157406 (6-(4-fluorophenyl)-2-[(8-methoxy-4-methyl-2-quinazolinyl)amino]-4(3H)-pyrimidinone) was identified as TCTP ligand with a KD of 0.87 ± 0.38. ZINC10157406 revealed growth inhibitory effects and caused G0/G1 cell cycle arrest in MOLT-4, SK-OV-3 and MCF-7 cells. ZINC10157406 (2 × IC50) downregulated TCTP expression by 86.70 ± 0.44% and upregulated p53 expression by 177.60 ± 12.46%. We validated ZINC10157406 binding to the p53 interaction site of TCTP and replacing p53 by co-immunoprecipitation. Discussion ZINC10157406 was identified as potent ligand of TCTP by in silico and in vitro methods. The compound bound to TCTP with a considerably higher affinity compared to artesunate as known TCTP inhibitor. We were able to demonstrate the effect of TCTP blockade at the p53 binding site, i.e. expression of TCTP decreased, whereas p53 expression increased. This effect was accompanied by a dose-dependent decrease of CDK2, CDK4, CDK, cyclin D1 and cyclin D3 causing a G0/G1 cell cycle arrest in MOLT-4, SK-OV-3 and MCF-7 cells. Our findings are supposed to stimulate further research on TCTP-specific small molecules for differentiation therapy in oncology.

scholarly journals STEM-20. A CANCER STEM CELL-SELECTIVE APOPTOSIS-INDUCING SMALL MOLECULE FOR THE TREATMENT OF GBM

2020 ◽  
Vol 22 (Supplement_2) ◽  
pp. ii200-ii200
Author(s):  
Stephen Skirboll ◽  
Natasha Lucki ◽  
Genaro Villa ◽  
Naja Vergani ◽  
Michael Bollong ◽  
...  
Keyword(s):  
Stem Cells ◽  
Cancer Stem Cells ◽  
Small Molecules ◽  
Small Molecule ◽  
Large Scale ◽  
Critical Role ◽  
Tumor Formation ◽  
Cell Type ◽  
Caspase 1 ◽  
Activation Assay

Abstract INTRODUCTION Glioblastoma multiforme (GBM) is the most aggressive form of primary brain cancer. A subpopulation of multipotent cells termed GBM cancer stem cells (CSCs) play a critical role in tumor initiation and maintenance, drug resistance, and recurrence following surgery. New therapeutic strategies for the treatment of GBM have recently focused on targeting CSCs. Here we have used an unbiased large-scale screening approach to identify drug-like small molecules that induce apoptosis in GBM CSCs in a cell type-selective manner. METHODS A luciferase-based survival assay of patient-derived GBM CSC lines was established to perform a large-scale screen of ∼one million drug-like small molecules with the goal of identifying novel compounds that are selectively toxic to chemoresistant GBM CSCs. Compounds found to kill GBM CSC lines as compared to control cell types were further characterized. A caspase activation assay was used to evaluate the mechanism of induced cell death. A xenograft animal model using patient-derived GBM CSCs was employed to test the leading candidate for suppression of in vivo tumor formation. RESULTS We identified a small molecule, termed RIPGBM, from the cell-based chemical screen that induces apoptosis in primary patient-derived GBM CSC cultures. The cell type-dependent selectivity of RIPGBM appears to arise at least in part from redox-dependent formation of a proapoptotic derivative, termed cRIPGBM, in GBM CSCs. cRIPGBM induces caspase 1-dependent apoptosis by binding to receptor-interacting protein kinase 2 (RIPK2) and acting as a molecular switch, which reduces the formation of a prosurvival RIPK2/TAK1 complex and increases the formation of a proapoptotic RIPK2/caspase 1 complex. In an intracranial GBM xenograft mouse model, RIPGBM was found to significantly suppress tumor formation. CONCLUSIONS Our chemical genetics-based approach has identified a small molecule drug candidate and a potential drug target that selectively targets cancer stem cells and provides an approach for the treatment of GBMs.

scholarly journals Forecast of International Trade of Lithium Carbonate Products in Importing Countries and Small-Scale Exporting Countries

2021 ◽  
Vol 13 (3) ◽  
pp. 1251
Author(s):  
Yichi Zhang ◽  
Zhiliang Dong ◽  
Sen Liu ◽  
Peixiang Jiang ◽  
Cuizhi Zhang ◽  
...  
Keyword(s):  
International Trade ◽  
Large Scale ◽  
Prediction Method ◽  
Lithium Ion ◽  
Lithium Carbonate ◽  
Raw Material ◽  
Small Scale ◽  
Trade Restrictions ◽  
Energy Field ◽  
Structure Of Trade

As the raw material of lithium-ion batteries, lithium carbonate plays an important role in the development of new energy field. Due to the extremely uneven distribution of lithium resources in the world, the security of supply in countries with less say would be greatly threatened if trade restrictions or other accidents occurred in large-scale exporting countries. It is of great significance to help these countries find new partners based on the existing trade topology. This study uses the link prediction method, based on the perspective of the topological structure of trade networks in various countries and trade rules, and eliminates the influence of large-scale lithium carbonate exporting countries on the lithium carbonate trade of other countries, to find potential lithium carbonate trade links among importing and small-scale exporting countries, and summarizes three trade rules: (1) in potential relationships involving two net importers, a relationship involving either China or the Netherlands is more likely to occur; (2) for all potential relationships, a relationship that actually occurred for more than two years in the period in 2009–2018 is more likely to occur in the future; and (3) potential relationships pairing a net exporter with a net importer are more likely to occur than other country combinations. The results show that over the next five to six years, Denmark and Italy, Netherlands and South Africa, Turkey and USA are most likely to have a lithium carbonate trading relationship, while Slovenia and USA, and Belgium and Thailand are the least likely to trade lithium carbonate. Through this study, we can strengthen the supply security of lithium carbonate resources in international trade, and provide international trade policy recommendations for the governments of importing countries and small-scale exporting countries.

scholarly journals An information theoretic approach to link prediction in multiplex networks

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Seyed Hossein Jafari ◽  
Amir Mahdi Abdolhosseini-Qomi ◽  
Masoud Asadpour ◽  
Maseud Rahgozar ◽  
Naser Yazdani
Keyword(s):  
Real World ◽  
Link Prediction ◽  
Large Scale ◽  
Similarity Measures ◽  
Prediction Method ◽  
General Purpose ◽  
Fast Method ◽  
Theoretic Approach ◽  
Multiplex Networks ◽  
Wide Range

AbstractThe entities of real-world networks are connected via different types of connections (i.e., layers). The task of link prediction in multiplex networks is about finding missing connections based on both intra-layer and inter-layer correlations. Our observations confirm that in a wide range of real-world multiplex networks, from social to biological and technological, a positive correlation exists between connection probability in one layer and similarity in other layers. Accordingly, a similarity-based automatic general-purpose multiplex link prediction method—SimBins—is devised that quantifies the amount of connection uncertainty based on observed inter-layer correlations in a multiplex network. Moreover, SimBins enhances the prediction quality in the target layer by incorporating the effect of link overlap across layers. Applying SimBins to various datasets from diverse domains, our findings indicate that SimBins outperforms the compared methods (both baseline and state-of-the-art methods) in most instances when predicting links. Furthermore, it is discussed that SimBins imposes minor computational overhead to the base similarity measures making it a potentially fast method, suitable for large-scale multiplex networks.

Measurements of Secondary Flows Downstream of a Turbine Cascade at Off-Design Incidence

2004 ◽  
Author(s):  
M. W. Benner ◽  
S. A. Sjolander ◽  
S. H. Moustapha
Keyword(s):  
Large Scale ◽  
Prediction Method ◽  
Leading Edge ◽  
Field Measurements ◽  
Inviscid Flow ◽  
Secondary Flows ◽  
Subsequent Paper ◽  
Empirical Prediction ◽  
Pressure Distributions ◽  
Passage Vortex

This paper presents experimental results of the secondary flows from two large-scale, low-speed, linear turbine cascades for which the incidence was varied. The aerofoils for the two cascades were designed for the same inlet and outlet conditions and differed mainly in their leading-edge geometries. Detailed flow field measurements were made upstream and downstream of the cascades and static pressure distributions were measured on the blade surfaces for three different values of incidence: 0, +10 and +20 degrees. The results from this experiment indicate that the strength of the passage vortex does not continue to increase with incidence, as would be expected from inviscid flow theory. The streamwise acceleration within the aerofoil passage seems to play an important role in influencing the strength of the vortex. The most recent off-design secondary loss correlation (Moustapha et al. [1]) includes leading-edge diameter as an influential correlating parameter. The correlation predicts that the secondary losses for the aerofoil with the larger leading-edge diameter are lower at off-design incidence; however, the opposite is observed experimentally. The loss results at high positive incidence have also high-lighted some serious shortcomings with the conventional method of loss decomposition. An empirical prediction method for secondary losses has been developed and will be presented in a subsequent paper.

Sign in / Sign up

Export Citation Format

Share Document