A machine learning-based quantitative model (LogBB_Pred) to predict the blood-brain barrier permeability (logBB value) of drug compounds.

MOTIVATION: Efficient assessment of the blood-brain barrier (BBB) penetration ability of a drug compound is one of the major hurdles in central nervous system drug discovery since experimental methods are costly and time-consuming. To advance and elevate the success rate of neurotherapeutic drug discovery, it is essential to develop an accurate computational quantitative model to determine the absolute logBB value (a logarithmic ratio of the concentration of a drug in the brain to its concentration in the blood) of a drug candidate. RESULTS: Here, we developed a quantitative model (LogBB_Pred) capable of predicting a logBB value of a query compound. The model achieved an R2 of 0.61 on an independent test dataset and outperformed other publicly available quantitative models. When compared with the available qualitative (classification) models that only classified whether a compound is BBB-permeable or not, our model achieved the same accuracy (0.85) with the best qualitative model and far-outperformed other qualitative models (accuracies between 0.64 and 0.70). For further evaluation, our model, quantitative models, and the qualitative models were evaluated on a real-world central nervous system drug screening library. Our model showed an accuracy of 0.97 while the other models showed an accuracy in the range of 0.29-0.83. Consequently, our model can accurately classify BBB-permeable compounds as well as predict the absolute logBB values of drug candidates. AVAILABILITY AND IMPLEMENTATION: Web server is freely available on the web at http://ssbio.cau.ac.kr/software/logbb_pred/. The data used in this study are available to download at http://ssbio.cau.ac.kr/software/logbb_pred/dataset.zip.

A multi-layered network model identifies Akt1 as a common modulator of neurodegeneration.

The accumulation of misfolded and aggregated proteins is a hallmark of neurodegenerative proteinopathies. Although multiple genetic loci have been associated with specific neurodegenerative diseases (NDs), molecular mechanisms that may have a broader relevance for most or all proteinopathies remain poorly resolved. In this study, we developed a multi-layered network expansion (MLnet) model to predict protein modifiers that are common to a group of diseases and, therefore, may have broader pathophysiological relevance for that group. When applied to the four NDs Alzheimer's disease (AD), Huntington's disease, and spinocerebellar ataxia types 1 and 3, we predicted multiple members of the insulin pathway, including PDK1, Akt1, InR, and sgg (GSK-3ß), as common modifiers. We validated these modifiers with the help of four Drosophila ND models. Further evaluation of Akt1 in human cell-based ND models revealed that activation of Akt1 signaling by the small molecule SC79 increased cell viability in all models. Moreover, treatment of AD model mice with SC79 enhanced their long-term memory and ameliorated dysregulated anxiety levels, which are commonly affected in AD patients. These findings validate MLnet as a valuable tool to uncover molecular pathways and proteins involved in the pathophysiology of entire disease groups and identify potential therapeutic targets that have relevance across disease boundaries. MLnet can be used for any group of diseases and is available as a web tool at http://ssbio.cau.ac.kr/software/mlnet.

Genotypes and phenotypes of DNM1 encephalopathy.

BACKGROUND: Variants in the dynamin-1 (DNM1) gene typically cause synaptopathy, leading to developmental and epileptic encephalopathy (DEE). We aimed to determine the genotypic and phenotypic spectrum of DNM1 encephalopathy beyond DEE. METHODS: Electroclinical phenotyping and genotyping of patients with a DNM1 variant were conducted for patients undergoing next-generation sequencing at our centre, followed by a systematic review. RESULTS: Six patients with heterozygous DNM1 variants were identified in our cohort. Three had a typical DEE phenotype characterised by epileptic spasms, tonic seizures and severe-to-profound intellectual disability with pathogenic variants located in the GTPase or middle domain. The other three patients had atypical phenotypes of milder cognitive impairment and focal epilepsy. Genotypically, two patients with atypical phenotypes had variants located in the GTPase domain, while the third patient had a novel variant (p.M648R) in the linker region between pleckstrin homology and GTPase effector domains. The third patient with an atypical phenotype showed normal development until he developed febrile status epilepticus. Our systematic review on 55 reported cases revealed that those with GTPase or middle domain variants had more severe intellectual disability (p<0.001) and lower functional levels of ambulation (p=0.001) or speech and language (p<0.001) than the rest. CONCLUSION: DNM1-related phenotypes encompass a wide spectrum of epilepsy and neurodevelopmental disorders, with specific variants underlying different phenotypes.

A computational clinical decision-supporting system to suggest effective anti-epileptic drugs for pediatric epilepsy patients based on deep learning models using patient's medical history.

BACKGROUND: Epilepsy, a chronic brain disorder characterized by abnormal brain activity that causes seizures and other symptoms, is typically treated using anti-epileptic drugs (AEDs) as the first-line therapy. However, due to the variations in their modes of action, identification of effective AEDs often relies on ad hoc trials, which is particularly challenging for pediatric patients. Thus, there is significant value in computational methods capable of assisting in the selection of AEDs, aiming to minimize unnecessary medication and improve treatment efficacy. RESULTS: In this study, we collected 7,507 medical records from 1,000 pediatric epilepsy patients and developed a computational clinical decision-supporting system for AED selection. This system leverages three multi-channel convolutional neural network (CNN) models tailored to three specific AEDs (vigabatrin, prednisolone, and clobazam). Each CNN model predicts whether a respective AED is effective on a given patient or not. The CNN models showed AUROCs of 0.90, 0.80, and 0.92 in 10-fold cross-validation, respectively. Evaluation on a hold-out test dataset further revealed positive predictive values (PPVs) of 0.92, 0.97, and 0.91 for the three respective CNN models, representing that suggested AEDs by our models would be effective in controlling epilepsy with a high accuracy and thereby reducing unnecessary medications for pediatric patients. CONCLUSION: Our CNN models in the system demonstrated high PPVs for the three AEDs, which signifies the potential of our approach to support the clinical decision-making by assisting doctors in recommending effective AEDs within the three AEDs for patients based on their medical history. This would result in a reduction in the number of unnecessary ad hoc attempts to find an effective AED for pediatric epilepsy patients.

A machine learning model for classifying G-protein-coupled receptors as agonists or antagonists.

BACKGROUND: G-protein coupled receptors (GPCRs) sense and transmit extracellular signals into the intracellular machinery by regulating G proteins. GPCR malfunctions are associated with a variety of signaling-related diseases, including cancer and diabetes; at least a third of the marketed drugs target GPCRs. Thus, characterization of their signaling and regulatory mechanisms is crucial for the development of effective drugs. RESULTS: In this study, we developed a machine learning model to identify GPCR agonists and antagonists. We designed two-step prediction models: the first model identified the ligands binding to GPCRs and the second model classified the ligands as agonists or antagonists. Using 990 selected subset features from 5270 molecular descriptors calculated from 4590 ligands deposited in two drug databases, our model classified non-ligands, agonists, and antagonists of GPCRs, and achieved an area under the ROC curve (AUC) of 0.795, sensitivity of 0.716, specificity of 0.744, and accuracy of 0.733. In addition, we verified that 70% (44 out of 63) of FDA-approved GPCR-targeting drugs were correctly classified into their respective groups. CONCLUSIONS: Studies of ligand-GPCR interaction recognition are important for the characterization of drug action mechanisms. Our GPCR-ligand interaction prediction model can be employed in the pharmaceutical sciences for the efficient virtual screening of putative GPCR-binding agonists and antagonists.

LightBBB: computational prediction model of blood-brain-barrier penetration based on LightGBM.

MOTIVATION: Identification of blood-brain barrier (BBB) permeability of a compound is a major challenge in neurotherapeutic drug discovery. Conventional approaches for BBB permeability measurement are expensive, time-consuming and labor-intensive. BBB permeability is associated with diverse chemical properties of compounds. However, BBB permeability prediction models have been developed using small datasets and limited features, which are usually not practical due to their low coverage of chemical diversity of compounds. Aim of this study is to develop a BBB permeability prediction model using a large dataset for practical applications. This model can be used for facilitated compound screening in the early stage of brain drug discovery. RESULTS: A dataset of 7162 compounds with BBB permeability (5453 BBB+ and 1709 BBB-) was compiled from the literature, where BBB+ and BBB- denote BBB-permeable and non-permeable compounds, respectively. We trained a machine learning model based on Light Gradient Boosting Machine (LightGBM) algorithm and achieved an overall accuracy of 89%, an area under the curve (AUC) of 0.93, specificity of 0.77 and sensitivity of 0.93, when 10-fold cross-validation was performed. The model was further evaluated using 74 central nerve system compounds (39 BBB+ and 35 BBB-) obtained from the literature and showed an accuracy of 90%, sensitivity of 0.85 and specificity of 0.94. Our model outperforms over existing BBB permeability prediction models. AVAILABILITYAND IMPLEMENTATION: The prediction server is available at http://ssbio.cau.ac.kr/software/bbb.

Advanced biotechnology using methyltransferase and its applications in bacteria: a mini review.

Since prokaryotic restriction-modification (RM) systems protect the host by cleaving foreign DNA by restriction endonucleases, it is difficult to introduce engineered plasmid DNAs into newly isolated microorganisms whose RM system is not discovered. The prokaryotes also possess methyltransferases to protect their own DNA from the endonucleases. As those methyltransferases can be utilized to methylate engineered plasmid DNAs before transformation and to enhance the stability within the cells, the study on methyltransferases in newly isolated bacteria is essential for genetic engineering. Here, we introduce the mechanism of the RM system, specifically the methyltransferases and their biotechnological applications. These biotechnological strategies could facilitate plasmid DNA-based genetic engineering in bacteria strains that strongly defend against foreign DNA.

Comparability of reference-based and reference-free transcriptome analysis approaches at the gene expression level.

BACKGROUND: Lately, high-throughput RNA sequencing has been extensively used to elucidate the transcriptome landscape and dynamics of cell types of different species. In particular, for most non-model organisms lacking complete reference genomes with high-quality annotation of genetic information, reference-free (RF) de novo transcriptome analyses, rather than reference-based (RB) approaches, are widely used, and RF analyses have substantially contributed toward understanding the mechanisms regulating key biological processes and functions. To date, numerous bioinformatics studies have been conducted for assessing the workflow, production rate, and completeness of transcriptome assemblies within and between RF and RB datasets. However, the degree of consistency and variability of results obtained by analyzing gene expression levels through these two different approaches have not been adequately documented. RESULTS: In the present study, we evaluated the differences in expression profiles obtained with RF and RB approaches and revealed that the former tends to be satisfactorily replaced by the latter with respect to transcriptome repertoires, as well as from a gene expression quantification perspective. In addition, we urge cautious interpretation of these findings. Several genes that are lowly expressed, have long coding sequences, or belong to large gene families must be validated carefully, whenever gene expression levels are calculated using the RF method. CONCLUSIONS: Our empirical results indicate important contributions toward addressing transcriptome-related biological questions in non-model organisms.

NFIB induces functional astrocytes from human pluripotent stem cell-derived neural precursor cells mimicking in vivo astrogliogenesis.

The ability to generate astrocytes from human pluripotent stem cells (hPSCs) offers a promising cellular model to study the development and physiology of human astrocytes. The extant methods for generating functional astrocytes required long culture periods and there remained much ambiguity on whether such paradigms follow the innate developmental program. In this report, we provided an efficient and rapid method for generating physiologically functional astrocytes from hPSCs. Overexpressing the nuclear factor IB in hPSC-derived neural precursor cells induced a highly enriched astrocyte population in 2 weeks. RNA sequencing and functional analyses demonstrated progressive transcriptomic and physiological changes in the cells, resembling in vivo astrocyte development. Further analyses substantiated previous results and established the MAPK pathway necessary for astrocyte differentiation. Hence, this differentiation paradigm provides a prospective in vitro model for human astrogliogenesis studies and the pathophysiology of neurological diseases concerning astrocytes.

The conserved microRNA miR-8-3p coordinates the expression of V-ATPase subunits to regulate ecdysone biosynthesis for Drosophila metamorphosis.

The steroid hormone ecdysone is the central regulator of insect metamorphosis, during which a growing, immature larva is remodeled, through pupal stages, to a reproductive adult. However, the underlying mechanisms of ecdysone-mediated metamorphosis remain to be fully elucidated. Here, we identified metamorphosis-associated microRNAs (miRNAs) and their potential targets by cross-linking immunoprecipitation coupled with deep sequencing of endogenous Argonaute 1 protein in Drosophila. Interestingly, miR-8-3p targeted five Vha genes encoding distinct subunits of vacuolar H+ -ATPase (V-ATPase), which has a vital role in the organellar acidification. The expression of ecdysone-responsive miR-8-3p is normally downregulated during Drosophila metamorphosis, but temporary overexpression of miR-8-3p in the whole body at the end of larval development led to defects in metamorphosis and survival, hallmarks of aberrant ecdysone signaling. In addition, miR-8-3p was expressed in the prothoracic gland (PG), which produces and releases ecdysone in response to prothoracicotropic hormone (PTTH). Notably, overexpression of miR-8-3p or knockdown of its Vha targets in the PG resulted in larger than normal, ecdysone-deficient larvae that failed to develop into the pupal stage but could be rescued by ecdysone feeding. Moreover, these animals showed defective PTTH signaling with a concomitant decrease in the expression of ecdysone biosynthetic genes. We also demonstrated that the regulatory network between the conserved miR-8-3p/miR-200 family and V-ATPase was functional in human cells. Consequently, our data indicate that the coordinated regulation of V-ATPase subunits by miR-8-3p is involved in Drosophila metamorphosis by controlling the ecdysone biosynthesis.

Recent Advances in Micro/Nanomaterial-Based Aptamer Selection Strategies.

Aptamers are artificial nucleic acid ligands that have been employed in various fundamental studies and applications, such as biological analyses, disease diagnostics, targeted therapeutics, and environmental pollutant detection. This review focuses on the recent advances in aptamer discovery strategies that have been used to detect various chemicals and biomolecules. Recent examples of the strategies discussed here are based on the classification of these micro/nanomaterial-mediated systematic evolution of ligands by exponential enrichment (SELEX) platforms into three categories: bead-mediated, carbon-based nanomaterial-mediated, and other nanoparticle-mediated strategies. In addition to describing the advantages and limitations of the aforementioned strategies, this review discusses potential strategies to develop high-performance aptamers.

2-D chemical structure image-based in silico model to predict agonist activity for androgen receptor.

BACKGROUND: Abnormal activation of human nuclear hormone receptors disrupts endocrine systems and thereby affects human health. There have been machine learning-based models to predict androgen receptor agonist activity. However, the models were constructed based on limited numerical features such as molecular descriptors and fingerprints. RESULT: In this study, instead of the numerical features, 2-D chemical structure images of compounds were used to build an androgen receptor toxicity prediction model. The images may provide unknown features that were not represented by conventional numerical features. As a result, the new strategy resulted in a construction of highly accurate prediction model: Mathews correlation coefficient (MCC) of 0.688, positive predictive value (PPV) of 0.933, sensitivity of 0.519, specificity of 0.998, and overall accuracy of 0.981 in 10-fold cross-validation. Validation on a test dataset showed MCC of 0.370, sensitivity of 0.211, specificity of 0.991, PPV of 0.882, and overall accuracy of 0.801. Our chemical image-based prediction model outperforms conventional models based on numerical features. CONCLUSION: Our constructed prediction model successfully classified molecular images into androgen receptor agonists or inactive compounds. The result indicates that 2-D molecular mimetic diagram would be used as another feature to construct molecular activity prediction models.

DNA steganography: hiding undetectable secret messages within the single nucleotide polymorphisms of a genome and detecting mutation-induced errors.

BACKGROUND: As cell engineering technology advances, more complex synthetically designed cells and metabolically engineered cells are being developed. Engineered cells are important resources in industry. Similar to image watermarking, engineered cells should be watermarked for protection against improper use. RESULTS: In this study, a DNA steganography methodology was developed to hide messages in variable regions (single nucleotide polymorphisms) of the genome to create hidden messages and thereby prevent from hacking. Additionally, to detect errors (mutations) within the encrypted messages, a block sum check algorithm was employed, similar to that used in network data transmission to detect noise-induced information changes. CONCLUSIONS: This DNA steganography methodology could be used to hide secret messages in a genome and detect errors within the encrypted messages. This approach is expected to be useful for tracking cells and protecting biological assets (e.g., engineered cells).

Cancer-Associated Fibroblasts Differentiated by Exosomes Isolated from Cancer Cells Promote Cancer Cell Invasion.

Cancer-associated fibroblasts (CAFs) in the cancer microenvironment play an essential role in metastasis. Differentiation of endothelial cells into CAFs is induced by cancer cell-derived exosomes secreted from cancer cells that transfer molecular signals to surrounding cells. Differentiated CAFs facilitate migration of cancer cells to different regions through promoting extracellular matrix (ECM) modifications. However, in vitro models in which endothelial cells exposed to cancer cell-derived exosomes secreted from various cancer cell types differentiate into CAFs or a microenvironmentally controlled model for investigating cancer cell invasion by CAFs have not yet been studied. In this study, we propose a three-dimensional in vitro cancer cell invasion model for real-time monitoring of the process of forming a cancer invasion site through CAFs induced by exosomes isolated from three types of cancer cell lines. The invasiveness of cancer cells with CAFs induced by cancer cell-derived exosomes (eCAFs) was significantly higher than that of CAFs induced by cancer cells (cCAFs) through physiological and genetic manner. In addition, different genetic tendencies of the invasion process were observed in the process of invading cancer cells according to CAFs. Our 3D microfluidic platform helps to identify specific interactions among multiple factors within the cancer microenvironment and provides a model for cancer drug development.

Computational determination of hERG-related cardiotoxicity of drug candidates.

BACKGROUND: Drug candidates often cause an unwanted blockage of the potassium ion channel of the human ether-a-go-go-related gene (hERG). The blockage leads to long QT syndrome (LQTS), which is a severe life-threatening cardiac side effect. Therefore, a virtual screening method to predict drug-induced hERG-related cardiotoxicity could facilitate drug discovery by filtering out toxic drug candidates. RESULT: In this study, we generated a reliable hERG-related cardiotoxicity dataset composed of 2130 compounds, which were carried out under constant conditions. Based on our dataset, we developed a computational hERG-related cardiotoxicity prediction model. The neural network model achieved an area under the receiver operating characteristic curve (AUC) of 0.764, with an accuracy of 90.1%, a Matthews correlation coefficient (MCC) of 0.368, a sensitivity of 0.321, and a specificity of 0.967, when ten-fold cross-validation was performed. The model was further evaluated using ten drug compounds tested on guinea pigs and showed an accuracy of 80.0%, an MCC of 0.655, a sensitivity of 0.600, and a specificity of 1.000, which were better than the performances of existing hERG-toxicity prediction models. CONCLUSION: The neural network model can predict hERG-related cardiotoxicity of chemical compounds with a high accuracy. Therefore, the model can be applied to virtual high-throughput screening for drug candidates that do not cause cardiotoxicity. The prediction tool is available as a web-tool at http://ssbio.cau.ac.kr/CardPred .

Artificial transformation methodologies for improving the efficiency of plasmid DNA transformation and simplifying its use.

The uptake of exogenous DNA materials through the cell membrane by bacteria, known as transformation, is essential for the genetic manipulation of bacteria and, thus, plays key roles in biotechnological and biological research. The efficiency of natural transformation is very low; therefore, various artificial transformation methods have been developed for simple and efficient bacterial transformation. The basic bacterial transformation method is based on chemical, physical, and electrical processes and other means to permeabilize the bacterial cell membrane to allow plasmid DNA uptake. With the introduction of novel chemicals, materials, and devices and the optimization of protocols, new transformation methods have become simpler, cheaper, and more reproducible for use in diverse bacterial species compared with conventional methods. In this review, artificial transformation methods have been classified according to the membrane-permeabilizing mechanisms employed by them. Their influential factors, transformation efficiency, advantages, disadvantages, and practical applications are briefly illustrated. Finally, physicochemical transformation as a new bacterial transformation technique has also been described.

In silico prediction of potential chemical reactions mediated by human enzymes.

BACKGROUND: Administered drugs are often converted into an ineffective or activated form by enzymes in our body. Conventional in silico prediction approaches focused on therapeutically important enzymes such as CYP450. However, there are more than thousands of different cellular enzymes that potentially convert administered drug into other forms. RESULT: We developed an in silico model to predict which of human enzymes including metabolic enzymes as well as CYP450 family can catalyze a given chemical compound. The prediction is based on the chemical and physical similarity between known enzyme substrates and a query chemical compound. Our in silico model was developed using multiple linear regression and the model showed high performance (AUC = 0.896) despite of the large number of enzymes. When evaluated on a test dataset, it also showed significantly high performance (AUC = 0.746). Interestingly, evaluation with literature data showed that our model can be used to predict not only enzymatic reactions but also drug conversion and enzyme inhibition. CONCLUSION: Our model was able to predict enzymatic reactions of a query molecule with a high accuracy. This may foster to discover new metabolic routes and to accelerate the computational development of drug candidates by enabling the prediction of the potential conversion of administered drugs into active or inactive forms.

MicroRNA signatures associated with thioacetamide-induced liver fibrosis in mice.

Multiple etiologies of liver injury are associated with fibrosis in which the key event is the activation of hepatic stellate cells (HSCs). Although microRNAs (miRNAs) are reportedly involved in fibrogenesis, the complete array of miRNA signatures associated with the disease has yet to be elucidated. Here, deep sequencing analysis revealed that compared to controls, 80 miRNAs were upregulated and 21 miRNAs were downregulated significantly in the thioacetamide (TAA)-induced mouse fibrotic liver. Interestingly, 58 of the upregulated miRNAs were localized to an oncogenic miRNA megacluster upregulated in liver cancer. Differential expression of some of the TAA-responsive miRNAs was confirmed, and their human orthologs were similarly deregulated in TGF-ß1-activated HSCs. Moreover, a functional analysis of the experimentally validated high-confidence miRNA targets revealed significant enrichment for the GO terms and KEGG pathways involved in HSC activation and liver fibrogenesis. This is the first comprehensive report of miRNAs profiles during TAA-induced mouse liver fibrosis.

Meta-analysis method for discovering reliable biomarkers by integrating statistical and biological approaches: An application to liver toxicity.

Biomarkers that are identified from a single study often appear to be biologically irrelevant or false positives. Meta-analysis techniques allow integrating data from multiple studies that are related but independent in order to identify biomarkers across multiple conditions. However, existing biomarker meta-analysis methods tend to be sensitive to the dataset being analyzed. Here, we propose a meta-analysis method, iMeta, which integrates t-statistic and fold change ratio for improved robustness. For evaluation of predictive performance of the biomarkers identified by iMeta, we compare our method with other meta-analysis methods. As a result, iMeta outperforms the other methods in terms of sensitivity and specificity, and especially shows robustness to study variance increase; it consistently shows higher classification accuracy on diverse datasets, while the performance of the others is highly affected by the dataset being analyzed. Application of iMeta to 59 drug-induced liver injury studies identified three key biomarker genes: Zwint, Abcc3, and Ppp1r3b. Experimental evaluation using RT-PCR and qRT-PCR shows that their expressional changes in response to drug toxicity are concordant with the result of our method. iMeta is available at http://imeta.kaist.ac.kr/index.html.

Changes in cellular microRNA expression induced by porcine circovirus type 2-encoded proteins.

Porcine circovirus type 2 (PCV2) is the primary causative agent of postweaning multisystemic wasting syndrome, which leads to serious economic losses in the pig industry worldwide. While the molecular basis of PCV2 replication and pathogenicity remains elusive, it is increasingly apparent that the microRNA (miRNA) pathway plays a key role in controlling virus-host interactions, in addition to a wide range of cellular processes. Here, we employed Solexa deep sequencing technology to determine which cellular miRNAs were differentially regulated after expression of each of three PCV2-encoded open reading frames (ORFs) in porcine kidney epithelial (PK15) cells. We identified 51 ORF1-regulated miRNAs, 74 ORF2-regulated miRNAs, and 32 ORF3-regulated miRNAs that differed in abundance compared to the control. Gene ontology analysis of the putative targets of these miRNAs identified transcriptional regulation as the most significantly enriched biological process, while KEGG pathway analysis revealed significant enrichment for several pathways including MAPK signaling, which is activated during PCV2 infection. Among the potential target genes of ORF-regulated miRNAs, two genes encoding proteins that are known to interact with PCV2-encoded proteins, zinc finger protein 265 (ZNF265) and regulator of G protein signaling 16 (RGS16), were selected for further analysis. We provide evidence that ZNF265 and RGS16 are direct targets of miR-139-5p and let-7e, respectively, which are both down-regulated by ORF2. Our data will initiate further studies to elucidate the roles of ORF-regulated cellular miRNAs in PCV2-host interactions.