The Promise of AI for DILI Prediction.
Ontology highlight
ABSTRACT:
SUBMITTER: Vall A
PROVIDER: S-EPMC8080874 | biostudies-literature | 2021
REPOSITORIES: biostudies-literature
Publications
The Promise of AI for DILI Prediction.
Vall Andreu A Sabnis Yogesh Y Shi Jiye J Class Reiner R Hochreiter Sepp S Klambauer Günter G
Frontiers in artificial intelligence 20210414
Drug-induced liver injury (DILI) is a common reason for the withdrawal of a drug from the market. Early assessment of DILI risk is an essential part of drug development, but it is rendered challenging prior to clinical trials by the complex factors that give rise to liver damage. Artificial intelligence (AI) approaches, particularly those building on machine learning, range from random forests to more recent techniques such as deep learning, and provide tools that can analyze chemical compounds ...[more]
Similar Datasets
Project description: Not available
| S-TOXR1052 | biostudies-other
Project description:BackgroundDrug-induced liver injury (DILI) is a major safety concern characterized by a complex and diverse pathogenesis. In order to identify DILI early in drug development, a better understanding of the injury and models with better predictivity are urgently needed. One approach in this regard are in silico models which aim at predicting the risk of DILI based on the compound structure. However, these models do not yet show sufficient predictive performance or interpretability to be useful for decision making by themselves, the former partially stemming from the underlying problem of labeling the in vivo DILI risk of compounds in a meaningful way for generating machine learning models.ResultsAs part of the Critical Assessment of Massive Data Analysis (CAMDA) "CMap Drug Safety Challenge" 2019 ( http://camda2019.bioinf.jku.at ), chemical structure-based models were generated using the binarized DILIrank annotations. Support Vector Machine (SVM) and Random Forest (RF) classifiers showed comparable performance to previously published models with a mean balanced accuracy over models generated using 5-fold LOCO-CV inside a 10-fold training scheme of 0.759 ± 0.027 when predicting an external test set. In the models which used predicted protein targets as compound descriptors, we identified the most information-rich proteins which agreed with the mechanisms of action and toxicity of nonsteroidal anti-inflammatory drugs (NSAIDs), one of the most important drug classes causing DILI, stress response via TP53 and biotransformation. In addition, we identified multiple proteins involved in xenobiotic metabolism which could be novel DILI-related off-targets, such as CLK1 and DYRK2. Moreover, we derived potential structural alerts for DILI with high precision, including furan and hydrazine derivatives; however, all derived alerts were present in approved drugs and were over specific indicating the need to consider quantitative variables such as dose.ConclusionUsing chemical structure-based descriptors such as structural fingerprints and predicted protein targets, DILI prediction models were built with a predictive performance comparable to previous literature. In addition, we derived insights on proteins and pathways statistically (and potentially causally) linked to DILI from these models and inferred new structural alerts related to this adverse endpoint.
| S-EPMC7814730 | biostudies-literature
Project description:Introduction Openness, conscientiousness, extroversion, agreeableness and neuroticism are dimensional personality traits known as the Big Five. Study attrition is a common but often hard to anticipate problem. Artificial intelligence (AI) could examine both fronts to mitigate the unpredictability of the latter. Objectives To investigate whether AI could predict study attrition employing personality traits scores. Methods Data from 2,697 questionnaires were analysed using an AI. The short form of the International Personality Item Pool was used to assess the Big Five personality traits on the first of three planned waves. The personality traits scores were employed to predict the missing of at least one wave. Overall attrition was 17.6%. The AI was conservatively tuned to minimize the negative likelihood ratio when confronting predicted and real attrition. The free and open source programming language R was used for all the analyses. Dataset source: Hansson, Isabelle; Berg, Anne Ingeborg; Thorvaldsson, Valgeir (2018), “Can personality predict longitudinal study attrition? Evidence from a population-based sample of older adults”, Mendeley Data, V1, doi: 10.17632/g3jx8zt2t9.1 Results Predictions obtained a negative likelihood ratio of 0.333 and a negative predictive value of 0.933. The results were indicative of fair performance. Conclusions AI might be useful to predict study retention. Furthermore, the results of this study might indicate a moderate effect of the Big Five on the probability of study retention. Finally, the AI used in this study is freely available, allowing anyone to experiment. Disclosure No significant relationships.
| S-EPMC9471583 | biostudies-literature
Project description:Drug discovery is a costly process which usually takes more than 10 years and billions of dollars for one successful drug to enter the market. Despite all the safety tests, drugs may still cause adverse reactions and be restricted in use or even withdrawn from the market. Drug-induced liver injury (DILI) is one of the major adverse drug reactions, and computational models may be used to predict and reduce it. To assess the computational prediction performance of DILI, we curated DILI endpoints from three databases and prepared drug features including chemical descriptors, therapeutic classifications, gene expressions, and binding proteins. We trained machine-learning models to predict the various DILI endpoints using different drug features. Using the optimal feature sets, the top-performing models obtained areas under the receiver operating characteristic curve (AUC) around 0.8 for some DILI endpoints. We found that some features, including therapeutic classifications and proteins, have good prediction performance towards DILI. We also discovered that the severity of DILI endpoints as well as the selection of negative samples may significantly affect the prediction results. Overall, our study provided a comprehensive collection, curation, and prediction of DILI endpoints using various drug features, which may help the drug researchers to better understand and prevent DILI during the drug discovery process.
| S-EPMC7254069 | biostudies-literature
Project description:Large language models (LLMs) like ChatGPT have potential applications in medical education such as helping students study for their licensing exams by discussing unclear questions with them. However, they require evaluation on these complex tasks. The purpose of this study was to evaluate how well publicly accessible LLMs performed on simulated UK medical board exam questions. 423 board-style questions from 9 UK exams (MRCS, MRCP, etc.) were answered by seven LLMs (ChatGPT-3.5, ChatGPT-4, Bard, Perplexity, Claude, Bing, Claude Instant). There were 406 multiple-choice, 13 true/false, and 4 "choose N" questions covering topics in surgery, pediatrics, and other disciplines. The accuracy of the output was graded. Statistics were used to analyze differences among LLMs. Leaked questions were excluded from the primary analysis. ChatGPT 4.0 scored (78.2%), Bing (67.2%), Claude (64.4%), and Claude Instant (62.9%). Perplexity scored the lowest (56.1%). Scores differed significantly between LLMs overall (p < 0.001) and in pairwise comparisons. All LLMs scored higher on multiple-choice vs true/false or "choose N" questions. LLMs demonstrated limitations in answering certain questions, indicating refinements needed before primary reliance in medical education. However, their expanding capabilities suggest a potential to improve training if thoughtfully implemented. Further research should explore specialty specific LLMs and optimal integration into medical curricula.
| S-EPMC11324724 | biostudies-literature
Project description:Growing evidence supports the critical role of tumour microenvironment (TME) in tumour progression, metastases, and treatment response. However, the in-situ interplay among various TME components, particularly between immune and tumour cells, are largely unknown, hindering our understanding of how tumour progresses and responds to treatment. While mainstream single-cell omics techniques allow deep, single-cell phenotyping, they lack crucial spatial information for in-situ cell-cell interaction analysis. On the other hand, tissue-based approaches such as hematoxylin and eosin and chromogenic immunohistochemistry staining can preserve the spatial information of TME components but are limited by their low-content staining. High-content spatial profiling technologies, termed spatial omics, have greatly advanced in the past decades to overcome these limitations. These technologies continue to emerge to include more molecular features (RNAs and/or proteins) and to enhance spatial resolution, opening new opportunities for discovering novel biological knowledge, biomarkers, and therapeutic targets. These advancements also spur the need for novel computational methods to mine useful TME insights from the increasing data complexity confounded by high molecular features and spatial resolution. In this review, we present state-of-the-art spatial omics technologies, their applications, major strengths, and limitations as well as the role of artificial intelligence (AI) in TME studies.
| S-EPMC10183599 | biostudies-literature
Project description:Introduction The dark triad is composed by the personality traits Machiavellianism, narcissism and psychopathy (MNP). Their complexity can make them difficult to interrelate. Artificial intelligence (AI) could help in this endeavour. Objectives To investigate whether AI could predict MNP from themselves. Methods Data from 210 questionnaires were analysed using an AI. The short Dark Triad questionnaire (SD3) was used to assess MNP. Two of the MNP scores were employed to predict the third one and the procedure was repeated for all of them alternatively. The AI was conservatively tuned to maximize the one-way random intraclass correlation coefficient (ICC) between predicted and real values. Pearson’s r was calculated too. The free and open source programming language R was used for all the analyses. Dataset source: Borráz-León, Javier I. (2020), “Dark triad, attractiveness, mate value, and sexual partners”, Mendeley Data, V1, doi: 10.17632/87vx6jfnrp.1 Results Machiavellianism, narcissism and psychopathy predictions obtained ICC of 0.593, 0.335, 0.505 and Pearson’s r of 0.608, 0.346, 0.548 respectively. The results were indicative of fair performance, mainly for Machiavellianism and psychopathy. Conclusions AI might be useful to predict MNP. This could be utile in many situations, such as dealing with missing data or deciding whether to formally test someone. Finally, the AI used in this study is freely available, allowing anyone to experiment. Disclosure No significant relationships.
| S-EPMC9471593 | biostudies-literature
Project description:Introduction SSRI-treated psychiatric disorders (STPD), such as general anxiety disorder and major depression disorder, are common psychiatric diagnoses. Serotonin-mediated effects of solar insolation are an active topic of research. Artificial intelligence (AI) could help to better examine that complex relation. Objectives To investigate whether AI could predict the STPD relying primarily on average ambient temperature and annual solar insolation. Methods Data of age, average ambient temperature and annual solar insolation were employed to predict STPD status in 7,587 subjects using an AI. To simplify the data analysis, only individuals with white ethnicity were assessed. SPTD prevalence was 17.1%. The AI was conservatively tuned to maximize the positive likelihood ratio considering predicted and real STPD statuses. The free and open source programming language R was used for all the analyses. Dataset source: Wortzel, Joshua; Kent, Shia; Avery, David; Al-Hamdan, Mohammad; Turner, Brandon; Norden, Justin; Norden, Michael; Haynor, David (2018), “Data for: Ambient temperature and solar insolation are associated with decreased prevalence of SSRI-treated psychiatric disorders”, Mendeley Data, V1, doi: 10.17632/trs43ybh92.1 Results Predictions obtained a positive likelihood ratio of 4.850. The results were indicative of fair performance. Conclusions AI might be useful to predict STPD. Furthermore, the results of this study might indicate a moderate effect of age, average ambient temperature and annual solar insolation on the probability of STPD occurrence. Finally, the AI used in this study is freely available, allowing anyone to experiment.
| S-EPMC9475708 | biostudies-literature
Project description:Introduction Openness, conscientiousness, extroversion, agreeableness and neuroticism are known as the Big Five personality traits (BFPT). They are theoretical building blocks of the personality and comprise wide and interconnected spectra. Artificial intelligence (AI) could help to grasp their complexity. Objectives To investigate whether AI could predict the BFPT from themselves. Methods Data from 2,697 questionnaires were analysed using an AI. The short form of the International Personality Item Pool was used to assess the BFPT. Four of the BFPT scores were employed to predict the fifth one and the procedure was repeated for all of them alternatively. The AI was conservatively tuned to maximize the one-way random intraclass correlation coefficient (ICC) between predicted and real values. Their Pearson’s r was calculated too. The free and open source programming language R was used for all the analyses. Dataset source: Hansson, Isabelle; Berg, Anne Ingeborg; Thorvaldsson, Valgeir (2018), “Can personality predict longitudinal study attrition? Evidence from a population-based sample of older adults”, Mendeley Data, V1, doi: 10.17632/g3jx8zt2t9.1 Results Openness, conscientiousness, extroversion, agreeableness and neuroticism predictions obtained ICC of 0.219, 0.146, 0.306, 0.354, 0.121 and Pearson’s r of 0.254, 0.149, 0.393, 0.446, 0.122 respectively. The results for extroversion and agreeableness were indicative of fair performance. Conclusions AI might be useful to predict personality traits, mainly extroversion and agreeableness. This could be utile in many situations, such as dealing with missing data or deciding whether to formally test someone. Finally, the AI used in this study is freely available, allowing anyone to experiment.
| S-EPMC9475767 | biostudies-literature