-
Briefings in Bioinformatics May 2024Accurate understanding of the biological functions of enzymes is vital for various tasks in both pathologies and industrial biotechnology. However, the existing methods...
Accurate understanding of the biological functions of enzymes is vital for various tasks in both pathologies and industrial biotechnology. However, the existing methods are usually not fast enough and lack explanations on the prediction results, which severely limits their real-world applications. Following our previous work, DEEPre, we propose a new interpretable and fast version (ifDEEPre) by designing novel self-guided attention and incorporating biological knowledge learned via large protein language models to accurately predict the commission numbers of enzymes and confirm their functions. Novel self-guided attention is designed to optimize the unique contributions of representations, automatically detecting key protein motifs to provide meaningful interpretations. Representations learned from raw protein sequences are strictly screened to improve the running speed of the framework, 50 times faster than DEEPre while requiring 12.89 times smaller storage space. Large language modules are incorporated to learn physical properties from hundreds of millions of proteins, extending biological knowledge of the whole network. Extensive experiments indicate that ifDEEPre outperforms all the current methods, achieving more than 14.22% larger F1-score on the NEW dataset. Furthermore, the trained ifDEEPre models accurately capture multi-level protein biological patterns and infer evolutionary trends of enzymes by taking only raw sequences without label information. Meanwhile, ifDEEPre predicts the evolutionary relationships between different yeast sub-species, which are highly consistent with the ground truth. Case studies indicate that ifDEEPre can detect key amino acid motifs, which have important implications for designing novel enzymes. A web server running ifDEEPre is available at https://proj.cse.cuhk.edu.hk/aihlab/ifdeepre/ to provide convenient services to the public. Meanwhile, ifDEEPre is freely available on GitHub at https://github.com/ml4bio/ifDEEPre/.
Topics: Deep Learning; Enzymes; Computational Biology; Software; Proteins; Databases, Protein; Algorithms
PubMed: 38942594
DOI: 10.1093/bib/bbae225 -
RMD Open Jun 2024Fibromyalgia (FM) is a complex disorder with widespread pain and emotional distress, posing diagnostic challenges. FM patients show altered cognitive and emotional...
BACKGROUND
Fibromyalgia (FM) is a complex disorder with widespread pain and emotional distress, posing diagnostic challenges. FM patients show altered cognitive and emotional processing, with a preferential allocation of attention to pain-related information. This attentional bias towards pain cues can impair cognitive functions such as inhibitory control, affecting patients' ability to manage and express emotions. Sentiment analysis using large language models (LLMs) can provide insights by detecting nuances in pain expression. This study investigated whether open-source LLM-driven sentiment analysis could aid FM diagnosis.
METHODS
40 patients with FM, according to the 2016 American College of Rheumatology Criteria and 40 non-FM chronic pain controls referred to rheumatology clinics, were enrolled. Transcribed responses to questions on pain and sleep were machine translated to English and analysed by the LLM Mistral-7B-Instruct-v0.2 using prompt engineering targeting FM-associated language nuances for pain expression ('prompt-engineered') or an approach without this targeting ('ablated'). Accuracy, precision, recall, specificity and area under the receiver operating characteristic curve (AUROC) were calculated using rheumatologist diagnosis as ground truth.
RESULTS
The prompt-engineered approach demonstrated accuracy of 0.87, precision of 0.92, recall of 0.84, specificity of 0.82 and AUROC of 0.86 for distinguishing FM. In comparison, the ablated approach had an accuracy of 0.76, precision of 0.75, recall of 0.77, specificity of 0.75 and AUROC of 0.76. The accuracy was superior to the ablated approach (McNemar's test p<0.001).
CONCLUSION
This proof-of-concept study suggests LLM-driven sentiment analysis, especially with prompt engineering, may facilitate FM diagnosis by detecting subtle differences in pain expression. Further validation is warranted, particularly the inclusion of secondary FM patients.
Topics: Humans; Fibromyalgia; Female; Middle Aged; Male; Adult; ROC Curve; Natural Language Processing; Language; Emotions; Aged; Chronic Pain
PubMed: 38942593
DOI: 10.1136/rmdopen-2024-004367 -
Progress in Molecular Biology and... 2024Respiratory infections such as Coronavirus disease 2019 are a substantial worldwide health challenge, frequently resulting in severe sickness and death, especially in... (Review)
Review
Respiratory infections such as Coronavirus disease 2019 are a substantial worldwide health challenge, frequently resulting in severe sickness and death, especially in susceptible groups. Conventional drug development for respiratory infections faces obstacles such as extended timescales, substantial expenses, and the rise of resistance to current treatments. Drug repurposing is a potential method that has evolved to quickly find and reuse existing medications for treating respiratory infections. Drug repurposing utilizes medications previously approved for different purposes, providing a cost-effective and time-efficient method to tackle pressing medical needs. This chapter summarizes current progress and obstacles in repurposing medications for respiratory infections, focusing on notable examples of repurposed pharmaceuticals and their probable modes of action. The text also explores the significance of computational approaches, high-throughput screening, and preclinical investigations in identifying potential candidates for repurposing. The text delves into the significance of regulatory factors, clinical trial structure, and actual data in confirming the effectiveness and safety of repurposed medications for respiratory infections. Drug repurposing is a valuable technique for quickly increasing the range of treatments for respiratory infections, leading to better patient outcomes and decreasing the worldwide disease burden.
Topics: Drug Repositioning; Humans; Respiratory Tract Infections; COVID-19 Drug Treatment; SARS-CoV-2; COVID-19; Antiviral Agents; Animals
PubMed: 38942538
DOI: 10.1016/bs.pmbts.2024.03.033 -
International Journal of Spine Surgery Jun 2024We provide a historical and technical perspective on the evolution of Kambin's triangle as a safe working corridor for percutaneous access to the intervertebral disc to...
We provide a historical and technical perspective on the evolution of Kambin's triangle as a safe working corridor for percutaneous access to the intervertebral disc to an anatomically expanded space to accommodate and facilitate open lumbar total joint replacement. The nearly 6-decade progression from intradiscal access in the intact lumbar spine to an enlarged working space following facetectomy to accommodate a transforaminal lumbar interbody fusion, and eventual further expansion via pedicle vertebral body osteotomy to support motion preservation with total joint replacement, represents a unique evolutionary pathway in surgical technique development. For each of these steps in evolution, we detail and provide the historical context of the corresponding surgical modifications required to expand the original anatomical boundaries of Kambin's triangle. It is postulated that the introduction of machine learning technologies coupled with innovations in robotics, materials science, and advanced imaging will further accelerate and refine the adaptation of more complex, precise, and efficacious surgical procedures to treat spinal degeneration via this working corridor.
PubMed: 38942442
DOI: 10.14444/8611 -
Neurobiology of Disease Jun 2024After ischemic stroke (IS), secondary injury is intimately linked to endoplasmic reticulum (ER) stress and body-brain crosstalk. Nonetheless, the underlying mechanism...
After ischemic stroke (IS), secondary injury is intimately linked to endoplasmic reticulum (ER) stress and body-brain crosstalk. Nonetheless, the underlying mechanism systemic immune disorder mediated ER stress in human IS remains unknown. In this study, 32 candidate ER stress-related genes (ERSRGs) were identified by overlapping MSigDB ER stress pathway genes and DEGs. Three Key ERSRGs (ATF6, DDIT3 and ERP29) were identified using LASSO, random forest, and SVM-RFE. IS patients with different ERSRGs profile were clustered into two groups using consensus clustering and the difference between 2 group was further explored by GSVA. Through immune cell infiltration deconvolution analysis, and middle cerebral artery occlusion (MCAO) mouse scRNA analysis, we found that the expression of 3 key ERSRGs were closely related with peripheral macrophage cell ER stress in IS and this was further confirmed by RT-qPCR experiment. These ERS genes might be helpful to further accurately regulate the central nervous system and systemic immune response through ER stress and have potential application value in clinical practice in IS.
PubMed: 38942324
DOI: 10.1016/j.nbd.2024.106583 -
The Science of the Total Environment Jun 2024Since the discovery of antibiotics, penicillin has remained the top choice in clinical medicine. With continuous advancements in biotechnology, penicillin production has... (Review)
Review
Since the discovery of antibiotics, penicillin has remained the top choice in clinical medicine. With continuous advancements in biotechnology, penicillin production has become cost-effective and efficient. Genetic engineering techniques have been employed to enhance biosynthetic pathways, leading to the production of new penicillin derivatives with improved properties and increased efficacy against antibiotic-resistant pathogens. Advances in bioreactor design, media formulation, and process optimization have contributed to higher yields, reduced production costs, and increased penicillin accessibility. While biotechnological advances have clearly benefited the global production of this life-saving drug, they have also created challenges in terms of waste management. Production fermentation broths from industries contain residual antibiotics, by-products, and other contaminants that pose direct environmental threats, while increased global consumption intensifies the risk of antimicrobial resistance in both the environment and living organisms. The current geographical and spatial distribution of antibiotic and penicillin consumption dramatically reveals a worldwide threat. These challenges are being addressed through the development of novel waste management techniques. Efforts are aimed at both upstream and downstream processing of antibiotic and penicillin production to minimize costs and improve yield efficiency while lowering the overall environmental impact. Yield optimization using artificial intelligence (AI), along with biological and chemical treatment of waste, is also being explored to reduce adverse impacts. The implementation of strict regulatory frameworks and guidelines is also essential to ensure proper management and disposal of penicillin production waste. This review is novel because it explores the key remaining challenges in antibiotic development, the scope of machine learning tools such as Quantitative Structure-Activity Relationship (QSAR) in modern biotechnology-driven production, improved waste management for antibiotics, discovering alternative path to reducing antibiotic use in agriculture through alternative meat production, addressing current practices, and offering effective recommendations.
PubMed: 38942308
DOI: 10.1016/j.scitotenv.2024.174236 -
The Science of the Total Environment Jun 2024Land subsidence in Bangkok, a pressing environmental challenge, demands sustained long-term policy interventions. Although mitigation measures have successfully...
Land subsidence in Bangkok, a pressing environmental challenge, demands sustained long-term policy interventions. Although mitigation measures have successfully alleviated subsidence rates within inner Bangkok, neighboring provinces continue to experience escalating rates. Conventional land-based monitoring methods exhibit limitations in coverage, and the anticipated nonlinear contributions of climatic and socioeconomic factors further complicate the spatiotemporal distribution of subsidence. This study aims to provide future subsidence predictions for the near (2023-2048), mid (2049-2074), and far-future (2075-2100), employing Interferometric Synthetic Aperture Radar (InSAR), Random Forest machine learning algorithm, and combined Shared Socioeconomic Pathways-Representative Concentration Pathways (SSP-RCPs) scenarios to address these challenges. The mean Line-of-Sight (LOS) velocity was found to be -7.0 mm/year, with a maximum of -53.5 mm/year recorded in Ayutthaya. The proposed model demonstrated good performance, yielding an R value of 0.84 and exhibiting no signs of overfitting. Across all scenarios, subsidence rates tend to increase by more than -9.0 mm/year in the near-future. However, for the mid and far-future, scenarios illustrate varying trends. The 'only-urban-LU change' scenario predicts a gradual recovery, while other change scenarios exhibit different tendencies.
PubMed: 38942307
DOI: 10.1016/j.scitotenv.2024.174285 -
Journal of Affective Disorders Jun 2024Although the effect sizes are modest, insomnia is consistently associated with suicidal thoughts and behaviors. Subgroup analyses can efficiently identify for whom...
BACKGROUND
Although the effect sizes are modest, insomnia is consistently associated with suicidal thoughts and behaviors. Subgroup analyses can efficiently identify for whom insomnia is most relevant to suicidal ideation. To improve clinical case identification, the present study sought to identify subclusters of lifetime suicidal ideators for whom insomnia was most closely related to current suicidal ideation.
METHODS
Data on N = 4750 lifetime suicidal ideators were extracted from the Military Suicide Research Consortium's Common Data Elements. Data on sociodemographic characteristics, severity and history of suicidal thoughts and behaviors, and related clinical characteristics were clustered by unsupervised machine learning algorithms. Robust Poisson regression estimated cluster by insomnia associations with current suicidal ideation.
RESULTS
Three clusters were identified: a modest symptom severity cluster (N = 1757, 37.0 %), an elevated severity cluster (N = 1444 30.4 %), and a high severity cluster (N = 1549 32.6 %). In Cluster 1, insomnia was associated with current suicidal ideation (PRR 1.29 [1.13-1.46]) and remained significant after adjusting for sociodemographic and clinical covariates. In Cluster 2, insomnia was associated with current suicidal ideation (PRR 1.14 [1.01-1.30]), but not after adjusting for sociodemographic and clinical covariates. In Cluster 3, insomnia was associated with current suicidal ideation (PRR 1.12 [1.03-1.21]) and remained significant after adjusting for sociodemographic covariates, but not clinical covariates.
LIMITATIONS
Cross-sectional design, lack of diagnostic data, non-representative sample.
CONCLUSION
Insomnia appears more closely related to current suicidal ideation among modest severity individuals than other subgroups. Future work should use prospective designs and more comprehensive risk factor measures to confirm these findings.
PubMed: 38942202
DOI: 10.1016/j.jad.2024.06.101 -
Survey of Ophthalmology Jun 2024Cystoid macular edema (CME) is a sight-threatening condition often associated with inflammatory and diabetic diseases. Early detection is crucial to prevent irreversible... (Review)
Review
Cystoid macular edema (CME) is a sight-threatening condition often associated with inflammatory and diabetic diseases. Early detection is crucial to prevent irreversible vision loss. Artificial intelligence (AI) has shown promise in automating CME diagnosis through optical coherence tomography (OCT) imaging, but its utility needs critical evaluation. This systematic review assesses the application of AI to diagnosis CME, specifically focusing on disorders like postoperative CME (Irvine Gass syndrome) and retinitis pigmentosa without obvious vasculopathy, using OCT imaging. A comprehensive search was conducted across 6 databases (PubMed, Scopus, Web of Science, Wiley, ScienceDirect, and IEEE) from 2018 to November, 2023. Twenty-three articles met the inclusion criteria and were selected for in-depth analysis. We evaluate AI's role in CME diagnosis and its performance in "detection", "classification" and "segmentation" of OCT retinal images. We found that convolutional neural network (CNN)-based methods consistently outperformed other machine learning techniques, achieving an average accuracy of over 96% in detecting and identifying CME from OCT images. Despite certain limitations such as dataset size and ethical concerns, the synergy between AI and OCT, particularly through CNNs, holds promise for significantly advancing CME diagnostics.
PubMed: 38942125
DOI: 10.1016/j.survophthal.2024.06.005 -
NeuroImage Jun 2024The prediction of Alzheimer's disease (AD) progression from its early stages is a research priority. In this context, the use of Artificial Intelligence (AI) in AD has...
BACKGROUND
The prediction of Alzheimer's disease (AD) progression from its early stages is a research priority. In this context, the use of Artificial Intelligence (AI) in AD has experienced a notable surge in recent years. However, existing investigations predominantly concentrate on distinguishing clinical phenotypes through cross-sectional approaches. This study aims to investigate the potential of modeling additional dimensions of the disease, such as variations in brain metabolism assessed via [F]-fluorodeoxyglucose positron emission tomography (FDG-PET), and utilize this information to identify patients with mild cognitive impairment (MCI) who will progress to dementia (pMCI).
METHODS
We analyzed data from 1,617 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) who had undergone at least one FDG-PET scan. We identified the brain regions with the most significant hypometabolism in AD and used Deep Learning (DL) models to predict future changes in brain metabolism. The best-performing model was then adapted under a multi-task learning framework to identify pMCI individuals. Finally, this model underwent further analysis using eXplainable AI (XAI) techniques.
RESULTS
Our results confirm a strong association between hypometabolism, disease progression, and cognitive decline. Furthermore, we demonstrated that integrating data on changes in brain metabolism during training enhanced the models' ability to detect pMCI individuals (sensitivity=88.4%, specificity=86.9%). Lastly, the application of XAI techniques enabled us to delve into the brain regions with the most significant impact on model predictions, highlighting the importance of the hippocampus, cingulate cortex, and some subcortical structures.
CONCLUSION
This study introduces a novel dimension to predictive modeling in AD, emphasizing the importance of projecting variations in brain metabolism under a multi-task learning paradigm.
PubMed: 38942101
DOI: 10.1016/j.neuroimage.2024.120695