APRILE: Exploring the Molecular Mechanisms of Drug Side Effects with Explainable Graph Neural Networks

Mapping Intimacies ◽

10.1101/2021.07.02.450937 ◽

2021 ◽

Author(s):

Hao Xu ◽

Shengqi Sang ◽

Herbert Yao ◽

Alexandra I. Herghelegiu ◽

Haiping Lu ◽

...

Keyword(s):

Neural Networks ◽

Side Effects ◽

Mental Disorders ◽

Side Effect ◽

Molecular Mechanisms ◽

Metabolic Diseases ◽

Biomedical Data ◽

Information Utilization ◽

Drug Pair ◽

Graph Neural Networks

With the majority of people 65 and over taking two or more medicines (polypharmacy), managing the side effects associated with polypharmacy is a global challenge. Explainable Artificial Intelligence (XAI) is necessary to reliably design safe polypharmacy. Here, we develop APRILE: a predictor-explainer framework based on graph neural networks to explore the molecular mechanisms underlying polypharmacy side effects by explaining predictions made by the predictors. For a side effect and its associated drug pair, or a set of side effects and their drug pairs, APRILE gives a set of proteins (drug targets or non-targets) and Gene Ontology (GO) items as the explanation. Using APRILE, we generate such explanations for 843,318 (learned) + 93,966 (novel) side effect--drug pair events, spanning 861 side effects (472 diseases, 485 symptoms and 9 mental disorders) and 20 disease categories. We show that our two new metrics, pharmacogenomic information utilization and protein-protein interaction information utilization, provide quantitative estimates of mechanism complexity. Explanations were significantly consistent with state of the art disease-gene associations for 232/239 (97%) side effects. Further, APRILE generated new insights into molecular mechanisms of four diverse categories of ADRs: infection, metabolic diseases, gastrointestinal diseases, and mental disorders, including paradoxical side effects. We demonstrate the viability of discovering polypharmacy side effect mechanisms by learning from an AI model trained on massive biomedical data. Consequently, it facilitates wider and more reliable use of AI in healthcare.

Download Full-text

Quality of life in schizophrenic patients

Dialogues in Clinical Neuroscience ◽

10.31887/dcns.2007.9.2/jbobes ◽

2007 ◽

Vol 9 (2) ◽

pp. 215-226 ◽

Cited By ~ 8

Keyword(s):

Quality Of Life ◽

Clinical Trials ◽

Depressive Symptoms ◽

Side Effects ◽

Mental Disorders ◽

Side Effect ◽

Antipsychotic Drugs ◽

Schizophrenic Patients ◽

The Impact

In the last decades, there has been increased interest in the field of quality of life in mental disorders in general, and particularly in schizophrenia. In addition, the appearance of the atypical antipsychotic drugs (amisulpride, aripiprazole, clozapine, olanzapine, quetiapine, risperidone, and ziprasidone) with different therapeutic and side-effect profiles, has promoted a greater interest in assessing the quality of life of schizophrenic patients. In this paper we will briefly summarize the difficulties in assessing quality of life in schizophrenic patients, as well as the results concerning their quality of life and the influence of psychopathology, especially negative and depressive symptoms, on it. We will also review data from recent clinical trials showing the impact ofantipsychotic treatments and their side effects upon quality of life.

Download Full-text

Cutaneous ulcers following hydroxyurea therapy

Phlebologie ◽

10.1055/s-0037-1621559 ◽

2004 ◽

Vol 33 (06) ◽

pp. 202-205 ◽

Cited By ~ 2

Author(s):

K. Hartmann ◽

S. Nagel ◽

T. Erichsen ◽

E. Rabe ◽

K. H. Grips ◽

...

Keyword(s):

Side Effects ◽

Side Effect ◽

Antineoplastic Agent ◽

Limited Time ◽

Myeloproliferative Diseases ◽

Dermatological Side Effects ◽

Chronic Myeloproliferative Diseases ◽

Hydroxyurea Therapy

SummaryHydroxyurea (HU) is usually a well tolerated antineoplastic agent and is commonly used in the treatment of chronic myeloproliferative diseases. Dermatological side effects are frequently seen in patients receiving longterm HU therapy. Cutaneous ulcers have been reported occasionally.We report on four patients with cutaneous ulcers whilst on long-term hydroxyurea therapy for myeloproliferative diseases. In all patients we were able to reduce the dose, or stop HU altogether and their ulcers markedly improved. Our observations suggest that cutaneous ulcers should be considered as possible side effect of long-term HU therapy and healing of the ulcers can be achieved not only by cessation of the HU treatment, but also by reducing the dose of hydroxyurea for a limited time.

Download Full-text

Graph Neural Networks for Prediction of Fuel Ignition Quality

10.26434/chemrxiv.12280325.v1 ◽

2020 ◽

Author(s):

Artur Schweidtmann ◽

Jan Rittig ◽

Andrea König ◽

Martin Grohe ◽

Alexander Mitsos ◽

...

Keyword(s):

Neural Networks ◽

Octane Number ◽

Molecular Graph ◽

Chemical Properties ◽

Graph Representation ◽

Structure Property ◽

Oxygenated Hydrocarbons ◽

Physico Chemical ◽

Ignition Quality ◽

Graph Neural Networks

<div>Prediction of combustion-related properties of (oxygenated) hydrocarbons is an important and challenging task for which quantitative structure-property relationship (QSPR) models are frequently employed. Recently, a machine learning method, graph neural networks (GNNs), has shown promising results for the prediction of structure-property relationships. GNNs utilize a graph representation of molecules, where atoms correspond to nodes and bonds to edges containing information about the molecular structure. More specifically, GNNs learn physico-chemical properties as a function of the molecular graph in a supervised learning setup using a backpropagation algorithm. This end-to-end learning approach eliminates the need for selection of molecular descriptors or structural groups, as it learns optimal fingerprints through graph convolutions and maps the fingerprints to the physico-chemical properties by deep learning. We develop GNN models for predicting three fuel ignition quality indicators, i.e., the derived cetane number (DCN), the research octane number (RON), and the motor octane number (MON), of oxygenated and non-oxygenated hydrocarbons. In light of limited experimental data in the order of hundreds, we propose a combination of multi-task learning, transfer learning, and ensemble learning. The results show competitive performance of the proposed GNN approach compared to state-of-the-art QSPR models making it a promising field for future research. The prediction tool is available via a web front-end at www.avt.rwth-aachen.de/gnn.</div>

Download Full-text

Conversational Emotion Recognition Using Self-Attention Mechanisms and Graph Neural Networks

10.21437/interspeech.2020-1703 ◽

2020 ◽

Author(s):

Zheng Lian ◽

Jianhua Tao ◽

Bin Liu ◽

Jian Huang ◽

Zhanlei Yang ◽

...

Keyword(s):

Neural Networks ◽

Emotion Recognition ◽

Graph Neural Networks

Download Full-text

Periodontal Pathogens and Neuropsychiatric Health

Current Topics in Medicinal Chemistry ◽

10.2174/1568026620666200110161105 ◽

2020 ◽

Vol 20 (15) ◽

pp. 1353-1397 ◽

Cited By ~ 3

Author(s):

Abhishek Wadhawan ◽

Mark A. Reynolds ◽

Hina Makkar ◽

Alison J. Scott ◽

Eileen Potocki ◽

...

Keyword(s):

Molecular Mechanisms ◽

Metabolic Diseases ◽

Low Grade ◽

Periodontal Pathogens ◽

Synergistic Interactions ◽

Brain Vasculature ◽

Micro Rnas ◽

Clinical Conditions ◽

Low Grade Inflammation ◽

Neuropsychiatric Illness

Increasing evidence incriminates low-grade inflammation in cardiovascular, metabolic diseases, and neuropsychiatric clinical conditions, all important causes of morbidity and mortality. One of the upstream and modifiable precipitants and perpetrators of inflammation is chronic periodontitis, a polymicrobial infection with Porphyromonas gingivalis (P. gingivalis) playing a central role in the disease pathogenesis. We review the association between P. gingivalis and cardiovascular, metabolic, and neuropsychiatric illness, and the molecular mechanisms potentially implicated in immune upregulation as well as downregulation induced by the pathogen. In addition to inflammation, translocation of the pathogens to the coronary and peripheral arteries, including brain vasculature, and gut and liver vasculature has important pathophysiological consequences. Distant effects via translocation rely on virulence factors of P. gingivalis such as gingipains, on its synergistic interactions with other pathogens, and on its capability to manipulate the immune system via several mechanisms, including its capacity to induce production of immune-downregulating micro-RNAs. Possible targets for intervention and drug development to manage distal consequences of infection with P. gingivalis are also reviewed.

Download Full-text

Ginger and Heart Health: From Mechanisms to Therapeutics

Current Molecular Pharmacology ◽

10.2174/1874467213666201209105005 ◽

2020 ◽

Vol 13 ◽

Author(s):

Sajad Fakhri ◽

Jayanta Kumar Patra ◽

Swagat Kumar Das ◽

Gitishree Das ◽

Mohammad Bagher Majnooni ◽

...

Keyword(s):

Side Effects ◽

Molecular Mechanisms ◽

Biological Activities ◽

Health Benefits ◽

Herbal Medicines ◽

Cardiovascular Complications ◽

National Database ◽

Heart Health ◽

Natural Entities ◽

Cardioprotective Effects

Background: As a major cause of morbidity and mortality, cardiovascular diseases (CVDs) are globally increasing. In spite of recent development in the management of cardiovascular complications, CVDs have remained a medical challenge. Numerous conventional drugs are used to play cardioprotective roles; however, they are associated with several side effects. Considering the rich phytochemistry and fewer side effects of herbal medicines, they have gained particular attention to develop novel herbal drugs with cardioprotective potentials. Amongst natural entities, ginger is an extensively used and well-known functional food and condiment, possessing plentiful bioactivities, like antiinflammatory, antioxidant, and antimicrobial properties in several disorders management. Objective: The current review deliberated phytochemical properties as well as the ginger/ginger constituents' biological activities and health benefits in several diseases, with particular attention to cardiovascular complications. Methods: A comprehensive research was conducted using multiple databases, including Scopus, PubMed, Medline, Web of Science, national database (Irandoc and SID), and related articles in terms of the health benefits and cardioprotective effects of ginger/ginger constituents. These data were collected from inception until August 2019. Results: In recent years, several herbal medicines were used to develop new drugs with more potency and also minor side effects. Amongst natural entities, ginger is an extensively used traditional medicine in several diseases. The crude extract, along with related pungent active constituents, is mostly attributed to heart health. The cardioprotective effects of ginger are contributed to its cardiotonic, antihypertensive, anti-hyperlipidemia, and anti-platelet effects. The signaling pathways and molecular mechanisms of ginger regarding its cardioprotective effects are also clarified. Conclusion: This study revealed the biological activities, health benefits, and cardioprotective properties of ginger/ginger constituents along with related mechanisms of action, which gave new insights to show new avenue in the treatment of CVDs.

Download Full-text