scholarly journals Chemical toxicity prediction based on semi-supervised learning and graph convolutional neural network

2021 ◽  
Vol 13 (1) ◽  
Author(s):  
Jiarui Chen ◽  
Yain-Whar Si ◽  
Chon-Wai Un ◽  
Shirley W. I. Siu
Keyword(s):  
Neural Network ◽  
Supervised Learning ◽  
Superior Performance ◽  
Chemical Toxicity ◽  
Animal Testing ◽  
Toxicity Prediction ◽  
Discovery Research ◽  
Drug Discovery Research

AbstractAs safety is one of the most important properties of drugs, chemical toxicology prediction has received increasing attentions in the drug discovery research. Traditionally, researchers rely on in vitro and in vivo experiments to test the toxicity of chemical compounds. However, not only are these experiments time consuming and costly, but experiments that involve animal testing are increasingly subject to ethical concerns. While traditional machine learning (ML) methods have been used in the field with some success, the limited availability of annotated toxicity data is the major hurdle for further improving model performance. Inspired by the success of semi-supervised learning (SSL) algorithms, we propose a Graph Convolution Neural Network (GCN) to predict chemical toxicity and trained the network by the Mean Teacher (MT) SSL algorithm. Using the Tox21 data, our optimal SSL-GCN models for predicting the twelve toxicological endpoints achieve an average ROC-AUC score of 0.757 in the test set, which is a 6% improvement over GCN models trained by supervised learning and conventional ML methods. Our SSL-GCN models also exhibit superior performance when compared to models constructed using the built-in DeepChem ML methods. This study demonstrates that SSL can increase the prediction power of models by learning from unannotated data. The optimal unannotated to annotated data ratio ranges between 1:1 and 4:1. This study demonstrates the success of SSL in chemical toxicity prediction; the same technique is expected to be beneficial to other chemical property prediction tasks by utilizing existing large chemical databases. Our optimal model SSL-GCN is hosted on an online server accessible through: https://app.cbbio.online/ssl-gcn/home.

scholarly journals Chemical Toxicity Prediction Based on Semi-supervised Learning and Graph Convolutional Neural Network

2021 ◽  
Author(s):  
Jiarui Chen ◽  
Yain-Whar Si ◽  
Chon-Wai Un ◽  
Shirley W. I. Siu
Keyword(s):  
Neural Network ◽  
Supervised Learning ◽  
Superior Performance ◽  
Chemical Toxicity ◽  
Animal Testing ◽  
Toxicity Prediction ◽  
Discovery Research ◽  
Drug Discovery Research

Abstract As safety is one of the most important properties of drugs, chemical toxicology prediction has received increasing attentions in the drug discovery research. Traditionally, researchers rely on in vitro and in vivo experiments to test the toxicity of chemical compounds. However, not only are these experiments time consuming and costly, but experiments that involve animal testing are increasingly subject to ethical concerns. While traditional machine learning (ML) methods have been used in the field with some success, the limited availability of annotated toxicity data is the major hurdle for further improving model performance. Inspired by the success of semi-supervised learning (SSL) algorithms, we propose a Graph Convolution Neural Network (GCN) to predict chemical toxicity and trained the network by the Mean Teacher (MT) SSL algorithm. Using the Tox21 data, our optimal SSL-GCN models for predicting the twelve toxicological endpoints achieve an average ROC-AUC score of 0.757 in the test set, which is a 6% improvement over GCN models trained by supervised learning and conventional ML methods. Our SSL-GCN models also exhibit superior performance when compared to models constructed using the built-in DeepChem ML methods. This study demonstrates that SSL can increase the prediction power of models by learning from unannotated data. The optimal unannotated to annotated data ratio ranges between 1:1 and 4:1. This study demonstrates the success of SSL in chemical toxicity prediction; the same technique is expected to be beneficial to other chemical property prediction tasks by utilizing existing large chemical databases.

scholarly journals In Vitro and In Vivo Antitrypanosomal Activities of Methanol Extract of Echinops kebericho Roots

2020 ◽  
Vol 2020 ◽  
pp. 1-6
Author(s):  
Debela Abdeta ◽  
Nigatu Kebede ◽  
Mirutse Giday ◽  
Getachew Terefe ◽  
Solomon Mequanente Abay
Keyword(s):  
Body Weight ◽  
Neglected Tropical Diseases ◽  
Field Isolate ◽  
Parasite Load ◽  
Phytochemical Analysis ◽  
Isolation And Purification ◽  
Discovery Research ◽  
Drug Discovery Research

Microbial resistance to the few conventional antitrypanosomal drugs, increasing resistance of vectors to insecticides, lack of effective vaccines, and adverse effects of the existing antitrypanosomal drugs justify the urgent need for effective, tolerable, and affordable drugs. We assessed antitrypanosomal effects of the hydromethanolic extract of Echinops kebericho Mesfin roots against Trypanosoma congolense field isolate using in vitro and in vivo techniques. Parasite load, packed cell volume (PCV), body weight, and rectal temperature in Swiss albino mice were assessed. This finding is part of the outcomes of drug discovery research for neglected tropical diseases. The extract arrested the motility of trypanosomes within 40 min at 4 and 2 mg/mL concentration, whereas in the untreated control, motility continued for more than 160 min. The extract also reduced parasitemia and prevented drop in PCV and body weight significantly (p<0.05), as compared to control. Phytochemical analysis showed the presence of flavonoids, triterpenes, steroids, saponins, glycosides, tannins, and alkaloids. It is observed that this extract has activity against the parasite. Isolation and purification of specific compounds are required to identify hit compounds responsible for the antitrypanosomal activity of the studied medicinal plant.

scholarly journals Pharmacologic Characterization of a Kinetic In Vitro Human Co-Culture Angiogenesis Model Using Clinically Relevant Compounds

2013 ◽  
Vol 18 (10) ◽  
pp. 1234-1245 ◽  
Author(s):  
Ashley Wolfe ◽  
Belinda O’Clair ◽  
Vincent E. Groppi ◽  
Dyke P. McEwen
Keyword(s):  
Growth Factor ◽  
Umbilical Vein ◽  
Dermal Fibroblasts ◽  
Tube Length ◽  
Endothelial Cell Proliferation ◽  
Discovery Research ◽  
Drug Discovery Research ◽  
Angiogenesis Model

Angiogenesis, the formation of new vessels from preexisting vessels, involves multiple cell types acting in concert to cause endothelial cell proliferation, migration, and differentiation into microvascular arrays. Under pathologic conditions, microenvironment changes result in altered blood vessel production. Historically, in vitro angiogenesis assays study individual aspects of the process and tend to be variable, difficult to quantify, and limited in clinical relevance. Here, we describe a kinetic, quantitative, co-culture angiogenesis model and demonstrate its relevance to in vivo pharmacology. Similar to in vivo angiogenesis, a co-culture of human umbilical vein endothelial cells with normal human dermal fibroblasts remains sensitive to multiple cytokines, resulting in a concentration-dependent stimulation of tube formation over time. Treatment with axitinib, a selective vascular endothelial growth factor (VEGF) antagonist, inhibited VEGF-mediated tube length and branch point formation and was selective for inhibiting VEGF over basic fibroblast growth factor (bFGF), similar to previous studies. Conversely, an FGFR-1 selective compound, PD-161570, was more potent at inhibiting bFGF-mediated angiogenesis. These results demonstrate the cytokine dynamics, selective pharmacology, and translational application of this model system. Finally, combining quantitative angiogenic biology with kinetic, live-content imaging highlights the importance of using validated in vitro models in drug discovery research.

Current Progress on the Genomics of Schistosomiasis for Drug Discovery and Diagnostics

2020 ◽  
Vol 20 (5) ◽  
pp. 598-610
Author(s):  
Nileshkumar Meghani ◽  
Beom-Jin Lee ◽  
Hardik Amin ◽  
Behzad Nili-Ahmadabadi ◽  
Saraswathy Nagendran
Keyword(s):  
Drug Discovery ◽  
Early Stage ◽  
Diagnostic Tools ◽  
Discovery Research ◽  
Drug Discovery Research ◽  
Tremendous Progress ◽  
Significant Research ◽  
Structural Insights

For a number of decades, schistosomiasis has remained a public threat and an economic burden in a number of countries, directly impacting over 200 million people. The past 15 years have seen tremendous progress in the development of high-throughput methods for targeting or compound selection that are vital to early-stage schistosome drug discovery research. Genomewide approaches to analyze gene expression at the transcriptional and other -omic levels have helped immensely for gaining insight into the pathways and mechanisms involved in the schistosomiasis and it is expected to revolutionize the drug discovery as well as related diagnostics. This review discusses the most recent progress of pharmacology and genomics concerning schistosomiasis with a focus on drug discovery and diagnostic tools. It also provides chemical structural insights of promising targets along with available in vitro and/or in vivo data. Although significant research has been done to identify new molecules for the treatment and new methods for diagnosis, the necessity of new options for the sustainable control of schistosomiasis remains a great challenge.

scholarly journals Preparation of Three-dimensional (3-D) Human Liver (HepaRG) Cultures for Histochemical and Immunohistochemical Staining and Light Microscopic Evaluation

2018 ◽  
Vol 46 (6) ◽  
pp. 653-659 ◽  
Author(s):  
Natasha P. Clayton ◽  
Alanna Burwell ◽  
Heather Jensen ◽  
Barbara F. Williams ◽  
Quashana D. Brown ◽  
...  
Keyword(s):  
Immunohistochemical Staining ◽  
Human Liver ◽  
Three Dimensional ◽  
Culture Systems ◽  
Discovery Research ◽  
Glass Slides ◽  
Drug Discovery Research ◽  
Microscopic Evaluation

The use of three-dimensional (3-D) in vitro culture systems (spheroids, organoids) in biomolecular and drug discovery research has become increasingly popular. The popularity is due, in part, to a diminished reliance on animal bioassays and a desire to develop physiologically relevant cell culture systems that simulate the in vivo tissue microenvironment. Most evaluations of 3-D cultures are by confocal microscopy and high-content imaging; however, these technologies do not allow for detailed cellular morphologic assessments or permit basic hematoxylin and eosin histologic evaluations. There are few studies that have reported detailed processes for preparing 3-D cultures for paraffin embedding and subsequent use for histochemical or immunohistochemical staining. In an attempt to do so, we have developed a protocol to paraffin-embed human liver spheroids that can be sectioned with a microtome and mounted onto glass slides for routine histochemical and immunohistochemical staining and light microscopic evaluations.

Identification of Therapeutically Active Molecules against Anthrax through Structure and Ligand based Drug Design

2019 ◽  
Vol 18 (27) ◽  
pp. 2294-2312
Author(s):  
Sisir Nandi ◽  
Mridula Saxena ◽  
Anil Kumar Saxena
Keyword(s):  
Drug Design ◽  
Protective Antigen ◽  
Structural Features ◽  
Anthrax Toxin ◽  
Biological Warfare ◽  
Discovery Research ◽  
Drug Discovery Research ◽  
Pathogenic Viruses

Background: People suffer from fatal diseases which are responsible for mortality. Potent devices and medicines are being developed to fight diseases caused by the microorganism for saving the lives of individuals. Highly pathogenic viruses and bacteria are being incorporated into biological warfare, which has become a major threat to mankind and causes the destruction of lives in a short span of time. Objective: The pathogen Bacillus anthracis, which is the causative of anthrax, is used in bioterrorism. Efforts are therefore being made to study the progress of biodefense drug discovery research in combating anthrax-based bioterrorism. Methods: This review describes the present status of the studies ontherapeutic measurement of anthrax toxin inhibitors towards inhibition of protective antigen, lethal and edema factors using chemometric and drug design tools to explore essential structural features for further design of active congeneric compounds. Results: The inhibitors estimated to show high activity through different models may be proposed for further synthesis and testing of biological activity in terms of anthrax toxin inhibition and cytotoxicity testing by in vitro and in vivo assays. Conclusion: Such an attempt is an insight of biodefense drug design against the dreadful threat to the nation due to anthrax-based terrorism and biological warfare.

scholarly journals Repurposing of the Open Access Malaria Box for Kinetoplastid Diseases Identifies Novel Active Scaffolds against Trypanosomatids

2015 ◽  
Vol 20 (5) ◽  
pp. 634-645 ◽  
Author(s):  
Marcel Kaiser ◽  
Louis Maes ◽  
Leela Pavan Tadoori ◽  
Thomas Spangenberg ◽  
Jean-Robert Ioset
Keyword(s):  
Drug Discovery ◽  
Leishmania Donovani ◽  
Human African Trypanosomiasis ◽  
Neglected Diseases ◽  
Discovery Research ◽  
Drug Discovery Research ◽  
Drug Pipeline ◽  
Malaria Box

Phenotypic screening had successfully been used for hit generation, especially in the field of neglected diseases, in which feeding the drug pipeline with new chemotypes remains a constant challenge. Here, we catalyze drug discovery research using a publicly available screening tool to boost drug discovery. The Malaria Box, assembled by the Medicines for Malaria Venture, is a structurally diverse set of 200 druglike and 200 probelike compounds distilled from more than 20,000 antimalarial hits from corporate and academic libraries. Repurposing such compounds has already identified new scaffolds against cryptosporidiosis and schistosomiasis. In addition to initiating new hit-to-lead activities, screening the Malaria Box against a plethora of other parasites would enable the community to better understand the similarities and differences between them. We describe the screening of the Malaria Box and triaging of the identified hits against kinetoplastids responsible for human African trypanosomiasis ( Trypanosoma brucei), Chagas disease ( Trypanosoma cruzi), and visceral leishmaniasis ( Leishmania donovani and Leishmania infantum). The in vitro and in vivo profiling of the most promising active compounds with respect to efficacy, toxicity, pharmacokinetics, and complementary druggable properties are presented and a collaborative model used as a way to accelerate the discovery process discussed.

scholarly journals In vitro and in vivo antitrypanosomal activities of Echinops kebericho root extract

2019 ◽  
Author(s):  
Debela Abdeta ◽  
Nigatu Kebede ◽  
Mirutse Giday ◽  
Getachew Terefe ◽  
Solomon Mequanente Abay
Keyword(s):  
Body Weight ◽  
Neglected Tropical Diseases ◽  
Field Isolate ◽  
Phytochemical Analysis ◽  
Root Extract ◽  
Isolation And Purification ◽  
Discovery Research ◽  
Drug Discovery Research

Abstract Objective: Microbial resistance to the few conventional antitrypanosomal drugs, increasing resistance of vectors to insecticides, lack of effective vaccines and adverse effects of the existing antitrypanosomal drugs justifies the urgent need for effective, tolerable and affordable drugs. We assessed antitrypanosomal effect of hydromethanolic extract of Echinops kebericho Mesfin roots against Trypanosoma congolense field isolate using in vitro and in vivo techniques. Parasite load, packed cell volume (PCV), body weight and rectal temperature in Swiss albino mice were assessed. This finding is part of the outcomes of drug discovery research for neglected tropical diseases. Result: The extract ceased motility of the trypanosomes within 40 min at 4 and 2 mg/ml concentration whereas in the untreated control motility continued for more than 160 min. The extract also reduced parasitemia, prevented drop in PCV and body weight significantly (p<0.05), as compared to control. Phytochemical analysis showed the presence of flavonoids, triterpines, steroids, saponins, glycosides, tannins and alkaloids. It is observed that this extract has activity against the parasite. Isolation and purification of specific compounds are required to identify hit compounds responsible for the antitrypanosomal activity of the studied medicinal plant.

Comparative Evaluation of In Vitro and In Vivo Assays for the Detection of Avian Infectious Bronchitis Virus as a Contaminant of Live Poultry Vaccines

1998 ◽  
Vol 26 (5) ◽  
pp. 629-634
Author(s):  
Emiliana Falcone ◽  
Edoardo Vignolo ◽  
Livia Di Trani ◽  
Simona Puzelli ◽  
Maria Tollis
Keyword(s):  
Infectious Bronchitis Virus ◽  
Serological Response ◽  
Avian Infectious Bronchitis Virus ◽  
Animal Testing ◽  
Rt Pcr ◽  
Pcr Assay ◽  
In Vivo Test ◽  
Infectious Bronchitis

A reverse transcriptase polymerase chain reaction (RT-PCR) assay specific for identifying avian infectious bronchitis virus (IBV) in poultry vaccines, and the serological response to IBV induced by the inoculation of chicks with a Newcastle disease vaccine spiked with the Massachusetts strain of IBV, were compared for their ability to detect IBV as a contaminant of avian vaccines. The sensitivity of the IBV-RT-PCR assay provided results which were at least equivalent to the biological effect produced by the inoculation of chicks, allowing this assay to be considered a valid alternative to animal testing in the quality control of avian immunologicals. This procedure can easily be adapted to detect a number of contaminants for which the in vivo test still represents the only available method of detection.

scholarly journals Homeostatic plasticity and external input shape neural network dynamics

2018 ◽  
Author(s):  
Johannes Zierenberg ◽  
Jens Wilting ◽  
Viola Priesemann
Keyword(s):  
Neural Network ◽  
Brain Area ◽  
External Input ◽  
Homeostatic Plasticity ◽  
Dynamic State ◽  
Network Topologies ◽  
Neural Network Dynamics ◽  
Spiking Activity

In vitro and in vivo spiking activity clearly differ. Whereas networks in vitro develop strong bursts separated by periods of very little spiking activity, in vivo cortical networks show continuous activity. This is puzzling considering that both networks presumably share similar single-neuron dynamics and plasticity rules. We propose that the defining difference between in vitro and in vivo dynamics is the strength of external input. In vitro, networks are virtually isolated, whereas in vivo every brain area receives continuous input. We analyze a model of spiking neurons in which the input strength, mediated by spike rate homeostasis, determines the characteristics of the dynamical state. In more detail, our analytical and numerical results on various network topologies show consistently that under increasing input, homeostatic plasticity generates distinct dynamic states, from bursting, to close-to-critical, reverberating and irregular states. This implies that the dynamic state of a neural network is not fixed but can readily adapt to the input strengths. Indeed, our results match experimental spike recordings in vitro and in vivo: the in vitro bursting behavior is consistent with a state generated by very low network input (< 0.1%), whereas in vivo activity suggests that on the order of 1% recorded spikes are input-driven, resulting in reverberating dynamics. Importantly, this predicts that one can abolish the ubiquitous bursts of in vitro preparations, and instead impose dynamics comparable to in vivo activity by exposing the system to weak long-term stimulation, thereby opening new paths to establish an in vivo-like assay in vitro for basic as well as neurological studies.

Sign in / Sign up

Export Citation Format

Share Document