Search

A novel CDK20 small molecule inhibitor discovered by artificial intelligence based on an AlphaFold-predicted structure demonstrates the first application of AlphaFold in hit identification for efficient drug discovery.

The search for radioprotectors is an ambitious goal with many practical applications. Particularly, the improvement of human radioresistance for space is an important task, which comes into view with the recent successes in the space industry. Currently, all radioprotective drugs can be divided into two large groups differing in their effectiveness depending on the type of exposure. The first of these is radioprotectors, highly effective for pulsed, and some types of relatively short exposure to irradiation. The second group consists of long-acting radioprotectors. These drugs are effective for prolonged and fractionated irradiation. They also protect against impulse exposure to ionizing radiation, but to a lesser extent than short-acting radioprotectors. Creating a database on radioprotectors is a necessity dictated by the modern development of science and technology. We have created an open database, Radioprotectors.org, containing an up-to-date list of substances with proven radioprotective properties. All radioprotectors are annotated with relevant chemical and biological information, including transcriptomic data, and can be filtered according to their properties. Additionally, the performed transcriptomics analysis has revealed specific transcriptomic profiles of radioprotectors, which should facilitate the search for potent radioprotectors.

Gallbladder cancer (GBC) is a rare malignancy, associated with poor disease prognosis with a 5-year survival of only 20%. This has been attributed to late presentation of the disease, lack of early diagnostic markers and limited efficacy of therapeutic interventions. Elucidation of molecular events in GBC can contribute to better management of the disease by aiding in the identification of therapeutic targets. To identify aberrantly activated signaling events in GBC, tandem mass tag-based quantitative phosphoproteomic analysis of five GBC cell lines was carried out. Proline-rich Akt substrate 40 kDa (PRAS40) was one of the proteins found to be hyperphosphorylated in all the invasive GBC cell lines. Tissue microarray-based immunohistochemical labeling of phospho-PRAS40 (T246) revealed moderate to strong staining in 77% of the primary gallbladder adenocarcinoma cases. Regulation of PRAS40 activity by inhibiting its upstream kinase PIM1 resulted in a significant decrease in cell proliferation, colony forming and invasive ability of GBC cells. Our results support the role of PRAS40 phosphorylation in GBC cell survival and aggressiveness. This study also elucidates phospho-PRAS40 as a clinical marker in GBC and the role of PIM1 as a therapeutic target in GBC. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s12079-018-00503-5) contains supplementary material, which is available to authorized users.

Here we present the application of deep neural network (DNN) ensembles trained on transcriptomic data to identify the novel markers associated with the mammalian embryonic-fetal transition (EFT). Molecular markers of this process could provide important insights into regulatory mechanisms of normal development, epimorphic tissue regeneration and cancer. Subsequent analysis of the most significant genes behind the DNNs classifier on an independent dataset of adult-derived and human embryonic stem cell (hESC)-derived progenitor cell lines led to the identification of COX7A1 gene as a potential EFT marker. COX7A1, encoding a cytochrome C oxidase subunit, was up-regulated in post-EFT murine and human cells including adult stem cells, but was not expressed in pre-EFT pluripotent embryonic stem cells or their in vitro-derived progeny. COX7A1 expression level was observed to be undetectable or low in multiple sarcoma and carcinoma cell lines as compared to normal controls. The knockout of the gene in mice led to a marked glycolytic shift reminiscent of the Warburg effect that occurs in cancer cells. The DNN approach facilitated the elucidation of a potentially new biomarker of cancer and pre-EFT cells, the embryo-onco phenotype, which may potentially be used as a target for controlling the embryonic-fetal transition.

High throughput technologies opened a new era in biomedicine by enabling massive analysis of gene expression at both RNA and protein levels. Unfortunately, expression data obtained in different experiments are often poorly compatible, even for the same biologic samples. Here, using experimental and bioinformatic investigation of major experimental platforms, we show that aggregation of gene expression data at the level of molecular pathways helps to diminish cross- and intra-platform bias otherwise clearly seen at the level of individual genes. We created a mathematical model of cumulative suppression of data variation that predicts the ideal parameters and the optimal size of a molecular pathway. We compared the abilities to aggregate experimental molecular data for the 5 alternative methods, also evaluated by their capacity to retain meaningful features of biologic samples. The bioinformatic method OncoFinder showed optimal performance in both tests and should be very useful for future cross-platform data analyses.

One of the major impediments in human aging research is the absence of a comprehensive and actionable set of biomarkers that may be targeted and measured to track the effectiveness of therapeutic interventions. In this study, we designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth, structure and optimization to predict human chronological age using a basic blood test. To train the DNNs, we used over 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex. The best performing DNN in the ensemble demonstrated 81.5 % epsilon-accuracy r = 0.90 with R(2) = 0.80 and MAE = 6.07 years in predicting chronological age within a 10 year frame, while the entire ensemble achieved 83.5% epsilon-accuracy r = 0.91 with R(2) = 0.82 and MAE = 5.55 years. The ensemble also identified the 5 most important markers for predicting human chronological age: albumin, glucose, alkaline phosphatase, urea and erythrocytes. To allow for public testing and evaluate real-life performance of the predictor, we developed an online system available at http://www.aging.ai. The ensemble approach may facilitate integration of multi-modal data linked to chronological age and sex that may lead to simple, minimally invasive, and affordable methods of tracking integrated biomarkers of aging in humans and performing cross-species feature importance analysis.

MicroRNAs (miRs) are short noncoding RNA molecules that regulate expression of target mRNAs. Many published sources provide information about miRs and their targets. However, bioinformatic tools elucidating higher level impact of the established total miR profiles, are still largely missing. Recently, we developed a method termed OncoFinder enabling quantification of the activities of intracellular molecular pathways basing on gene expression data. Here we propose a new technique, MiRImpact, which enables to link miR expression data with its estimated outcome on the regulation of molecular pathways, like signaling, metabolic, cytoskeleton rearrangement, and DNA repair pathways. MiRImpact uses OncoFinder rationale for pathway activity calculations, with the major distinctions that (i) it deals with the concentrations of miRs - known regulators of gene products participating in molecular pathways, and (ii) miRs are considered as negative regulators of target molecules, if other is not specified. MiRImpact operates with 2 types of databases: for molecular targets of miRs and for gene products participating in molecular pathways. We applied MiRImpact to compare regulation of human bladder cancer-specific signaling pathways at the levels of mRNA and miR expression. We took 2 most complete alternative databases of experimentally validated miR targets – miRTarBase and DianaTarBase, and an OncoFinder database featuring 2725 gene products and 271 signaling pathways. We showed that the impact of miRs is orthogonal to pathway regulation at the mRNA level, which stresses the importance of studying posttranscriptional regulation of gene expression. We also report characteristic set of miR and mRNA regulation features linked with bladder cancer.

Here we present the application of deep neural network (DNN) ensembles trained on transcriptomic data to identify the novel markers associated with the mammalian embryonic-fetal transition (EFT). Molecular markers of this process could provide important insights into regulatory mechanisms of normal development, epimorphic tissue regeneration and cancer. Subsequent analysis of the most significant genes behind the DNNs classifier on an independent dataset of adult-derived and human embryonic stem cell (hESC)-derived progenitor cell lines led to the identification of COX7A1 gene as a potential EFT marker. COX7A1, encoding a cytochrome C oxidase subunit, was up-regulated in post-EFT murine and human cells including adult stem cells, but was not expressed in pre-EFT pluripotent embryonic stem cells or their in vitro-derived progeny. COX7A1 expression level was observed to be undetectable or low in multiple sarcoma and carcinoma cell lines as compared to normal controls. The knockout of the gene in mice led to a marked glycolytic shift reminiscent of the Warburg effect that occurs in cancer cells. The DNN approach facilitated the elucidation of a potentially new biomarker of cancer and pre-EFT cells, the embryo-onco phenotype, which may potentially be used as a target for controlling the embryonic-fetal transition.

// Ksenia Lezhnina 1,2 , Olga Kovalchuk 3,4 , Alexander A. Zhavoronkov, 2,5,6 , Mikhail B. Korzinkin 1 , Anastasia A. Zabolotneva 7 , Peter V. Shegay 8 , Dmitry G. Sokov 9 , Nurshat M. Gaifullin 10,11 , Igor G. Rusakov 8 , Alexander M. Aliper 1,2 , Sergey A. Roumiantsev 2 , Boris Y. Alekseev 8 , Nikolay M. Borisov 12 and Anton A. Buzdin 1,2,7 1 Pathway Pharmaceuticals, Wan Chai, Hong Kong, Hong Kong SAR 2 Laboratory of Bioinformatics, D. Rogachyov Federal Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russia 3 Department of Biological Sciences, University of Lethbridge, 4401 University Drive, Lethbridge, AB, T1K 3M4 4 Canada Cancer and Aging Research Laboratories, Lethbridge, AB, Canada 5 Insilico Medicine, Inc, ETC, Johns Hopkins University, Baltimore, MD 6 Faculty of Biological and Medical Physics, Moscow Institute of Physics and Technology 7 Group for Genomic Regulation of Cell Signaling Systems, Shemyakn-Ovchinnikov Institute of Bioorganic Chemistry, Moscow, Russia 8 P.A. Herzen Moscow Oncological Research Institute, Moscow, Russia 9 Moscow 1st Oncological Hospital, Moscow, Russia 10 Lomonosov Moscow State University, Faculty of Fundamental Medicine, Moscow, Russia 11 Russian medical postgraduate academy , Moscow, Russia 12 Laboratory of Systems Biology, A.I. Burnasyan Federal Medical Biophysical Center, Moscow, Russia Correspondence: Anton A. Buzdin, email: // Keywords : Bladder cancer, Intracellular signaling pathway activation, Gene expression, Transcriptome profiling, Molecular markers, AUC Received : August 21, 2014 Accepted : September 15, 2014 Published : September 16, 2014 Abstract We recently proposed a new bioinformatic algorithm called OncoFinder for quantifying the activation of intracellular signaling pathways. It was proved advantageous for minimizing errors of high-throughput gene expression analyses and showed strong potential for identifying new biomarkers. Here, for the first time, we applied OncoFinder for normal and cancerous tissues of the human bladder to identify biomarkers of bladder cancer. Using Illumina HT12v4 microarrays, we profiled gene expression in 17 cancer and seven non-cancerous bladder tissue samples. These experiments were done in two independent laboratories located in Russia and Canada. We calculated pathway activation strength values for the investigated transcriptomes and identified signaling pathways that were regulated differently in bladder cancer (BC) tissues compared with normal controls. We found, for both experimental datasets, 44 signaling pathways that serve as excellent new biomarkers of BC, supported by high area under the curve (AUC) values. We conclude that the OncoFinder approach is highly efficient in finding new biomarkers for cancer. These markers are mathematical functions involving multiple gene products, which distinguishes them from “traditional” expression biomarkers that only assess concentrations of single genes.

High throughput technologies opened a new era in biomedicine by enabling massive analysis of gene expression at both RNA and protein levels. Unfortunately, expression data obtained in different experiments are often poorly compatible, even for the same biological samples. Here, using experimental and bioinformatic investigation of major experimental platforms, we show that aggregation of gene expression data at the level of molecular pathways helps to diminish cross- and intra-platform bias otherwise clearly seen at the level of individual genes. We created a mathematical model of cumulative suppression of data variation that predicts the ideal parameters and the optimal size of a molecular pathway. We compared the abilities to aggregate experimental molecular data for the five alternative methods, also evaluated by their capacity to retain meaningful features of biological samples. The bioinformatic method OncoFinder showed optimal performance in both tests and should be very useful for future cross-platform data analyses.

// Denis Shepelin 1, 2 , Mikhail Korzinkin 1, 3 , Anna Vanyushina 4 , Alexander Aliper 4 , Nicolas Borisov 3, 5 , Raif Vasilov 5 , Nikolay Zhukov 3, 6 , Dmitry Sokov 7 , Vladimir Prassolov 8 , Nurshat Gaifullin 9 , Alex Zhavoronkov 10 , Bhupinder Bhullar 11 , Anton Buzdin 1, 4, 5 1 Pathway Pharmaceuticals, Wan Chai, Hong Kong, Hong Kong SAR 2 Group for Genomic Analysis of Cell Signaling Systems, Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Moscow, Russia 3 First Oncology Research and Advisory Center, Moscow, Russia 4 Laboratory of Bioinformatics, D. Rogachyov Federal Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russia 5 National Research Centre “Kurchatov Institute”, Centre for Convergence of Nano-, Bio-, Information and Cognitive Sciences and Technologies, Moscow, Russia 6 Pirogov Russian National Research Medical University, Department of Oncology, Hematology and Radiotherapy, Moscow, Russia 7 Moscow 1st Oncological Hospital, Moscow Russia 8 Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Mosow, Russia 9 Moscow State University, Faculty of Fundamental Medicine, Moscow, Russia 10 Insilico Medicine, Inc, ETC, Johns Hopkins University, Baltimore, MD, USA 11 Novartis Institute for Biomedical Research, Basel, Switzerland Correspondence to: Anton Buzdin, e-mail: buzdin@ponkc.com Keywords: transition from nevus to primary and metastatic melanoma, OncoFinder, intracellular molecular networks, metabolic and signaling pathways, machine learning algorithms Received: January 17, 2015 Accepted: November 11, 2015 Published: November 26, 2015 ABSTRACT Melanoma is the most aggressive and dangerous type of skin cancer, but its molecular mechanisms remain largely unclear. For transcriptomic data of 478 primary and metastatic melanoma, nevi and normal skin samples, we performed high-throughput analysis of intracellular molecular networks including 592 signaling and metabolic pathways. We showed that at the molecular pathway level, the formation of nevi largely resembles transition from normal skin to primary melanoma. Using a combination of bioinformatic machine learning algorithms, we identified 44 characteristic signaling and metabolic pathways connected with the formation of nevi, development of primary melanoma, and its metastases. We created a model describing formation and progression of melanoma at the level of molecular pathway activation. We discovered six novel associations between activation of metabolic molecular pathways and progression of melanoma: for allopregnanolone biosynthesis, L-carnitine biosynthesis, zymosterol biosynthesis (inhibited in melanoma), fructose 2, 6-bisphosphate synthesis and dephosphorylation, resolvin D biosynthesis (activated in melanoma), D-myo-inositol hexakisphosphate biosynthesis (activated in primary, inhibited in metastatic melanoma). Finally, we discovered fourteen tightly coordinated functional clusters of molecular pathways. This study helps to decode molecular mechanisms underlying the development of melanoma.

// Nikolay M. Borisov 1, 2 , Nadezhda V. Terekhanova 1, 3 , Alexander M. Aliper 1, 3 , Larisa S. Venkova 1, 2 , Philip Yu Smirnov 1, 2 , Sergey Roumiantsev 1, 3 , Mikhail B. Korzinkin 1, 2 , Alex A. Zhavoronkov 1, 3, 4 , Anton A. Buzdin 1, 3, 4 1 Pathway Pharmaceuticals, Wan Chai, Hong Kong, Hong Kong SAR 2 Laboratory of Systems Biology, A.I. Burnasyan Federal Medical Biophysical Center, Moscow, 123182, Russia 3 Laboratory of Bioinformatics, D. Rogachyov Federal Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, 117198, Russia 4 Group for Genomic Regulation of Cell Signaling Systems, Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Moscow, 117997, Russia Correspondence to: Anton Buzdin, e-mail: bu3din@mail.ru Key words: Cancer, Intracellular signaling pathway activation, Gene expression, Transcriptome profiling, Molecular markers, AUC Received: July 16, 2014 Accepted: August 16, 2014 Published: October 13, 2014 ABSTRACT Identification of reliable and accurate molecular markers remains one of the major challenges of contemporary biomedicine. We developed a new bioinformatic technique termed OncoFinder that for the first time enables to quantatively measure activation of intracellular signaling pathways basing on transcriptomic data. Signaling pathways regulate all major cellular events in health and disease. Here, we showed that the Pathway Activation Strength (PAS) value itself may serve as the biomarker for cancer, and compared it with the “traditional” molecular markers based on the expression of individual genes. We applied OncoFinder to profile gene expression datasets for the nine human cancer types including bladder cancer, basal cell carcinoma, glioblastoma, hepatocellular carcinoma, lung adenocarcinoma, oral tongue squamous cell carcinoma, primary melanoma, prostate cancer and renal cancer, totally 292 cancer and 128 normal tissue samples taken from the Gene expression omnibus (GEO) repository. We profiled activation of 82 signaling pathways that involve ~2700 gene products. For 9/9 of the cancer types tested, the PAS values showed better area-under-the-curve (AUC) scores compared to the individual genes enclosing each of the pathways. These results evidence that the PAS values can be used as a new type of cancer biomarkers, superior to the traditional gene expression biomarkers.