Search

PURPOSE: To compare the evaluation metrics for deep learning methods that were developed using imbalanced imaging data in osteoarthritis studies. MATERIALS AND METHODS: This retrospective study utilized 2996 sagittal intermediate-weighted fat-suppressed knee MRIs with MRI Osteoarthritis Knee Score readings from 2467 participants in the Osteoarthritis Initiative study. We obtained probabilities of the presence of bone marrow lesions (BMLs) from MRIs in the testing dataset at the sub-region (15 sub-regions), compartment, and whole-knee levels based on the trained deep learning models. We compared different evaluation metrics (e.g., receiver operating characteristic (ROC) and precision-recall (PR) curves) in the testing dataset with various class ratios (presence of BMLs vs. absence of BMLs) at these three data levels to assess the models performance. RESULTS: In a subregion with an extremely high imbalance ratio, the model achieved a ROC-AUC of 0.84, a PR-AUC of 0.10, a sensitivity of 0, and a specificity of 1. CONCLUSION: The commonly used ROC curve is not sufficiently informative, especially in the case of imbalanced data. We provide the following practical suggestions based on our data analysis: 1) ROC-AUC is recommended for balanced data, 2) PR-AUC should be used for moderately imbalanced data (i.e., when the proportion of the minor class is above 5% and less than 50%), and 3) for severely imbalanced data (i.e., when the proportion of the minor class is below 5%), it is not practical to apply a deep learning model, even with the application of techniques addressing imbalanced data issues.

Abstract Background Elevation of carpal tunnel pressure is known to be associated with carpal tunnel syndrome. This study aimed to correlate the shear wave elastography in the transverse carpal ligament (TCL) with carpal tunnel pressures using a cadaveric model. Methods Eight human cadaveric hands were dissected to evacuate the tunnels. A medical balloon was inserted into each tunnel and connected to a pressure regulator to simulate tunnel pressure in the range of 0-210 mmHg with an increment of 30 mmHg. Shear wave velocity and modulus was measure in the middle of TCL. Results SWV and SWE were significantly dependent on the pressure levels (p

Background: SZ-685C, an anthracycline compound derived from the mangrove endophytic fungus Halorosellinia sp. (No. 1403) collected from the South China Sea, has shown strong anticancer activities. Non-functioning pituitary adenomas (NFPAs) are a type of tumor that can be challenging to manage clinically and have a significant unmet medical need. Our research has found that SZ-685C showed an inhibitory effect on the viability, migration ability, and proliferation ability of a human non-functioning pituitary tumor-derived folliculostellate (PDFS) cell line. Methods: SZ-685C was prepared and purified from the mangrove endophytic fungus No. 1403. PDFS cells were exposed to SZ-685C, and the effect of SZ-685C on PDFS cells was evaluated. RNA sequencing was used to analyze the miRNA expression profile in PDFS cells of the control group and SZ-685C-treated group. Quantitative polymerase chain reaction (qPCR) was performed to verify the expression of selected miR-340-3p. The effects of SZ-685C on PDFS cells after overexpression of miR-340-3p were evaluated. Dual-luciferase reporter assays showed PPP1CB is a direct target of miR-340-3p. Finally, the action pathway of the selected miR-340-3p was predicted and evaluated through bioinformatics analysis. Results: SZ-685C reduced cell viability in PDFS cells, accompanied by inhibition of migration ability and proliferation ability. The IC50 value for 24 h is 9.144 ± 0.991 μM, and for 48 h is 4.635 ± 0.551 μM. SZ-685C increased the protein levels of Beclin 1, the ratio of LC3-II to LC3-I, and LAMP-1, and down-regulated p62. MiRNA sequencing and further validation showed that miR-340-3p significantly decreased in PDFS cells treated with SZ-685C. After overexpression of miR-340-3p, the inhibition of viability, migration ability, proliferation ability, and autophagy-promoting effect of SZ-685C on PDFS cells were weakened. SZ-685C caused a decrease in PPP1CB expression and activation of the ERK pathway in PDFS cells, and this trend was reversed after overexpression of miR-340-3p. Conclusions: SZ-685C downregulates the expression of miR-340-3p in PDFS cells, thereby reducing the expression of PPP1CB and activating the ERK pathway to promote autophagic cell death, leading to inhibition of PDFS cell growth.

Abstract Background The plant hormone auxin is widely involved in plant growth, development, and morphogenesis, and the TIR1/AFB and AUX/IAA proteins are closely linked to rapid auxin response and signal transmission. However, their evolutionary history, historical patterns of expansion and contraction, and changes in interaction relationships are still unknown. Results Here, we analyzed the gene duplications, interactions, and expression patterns of TIR1/AFBs and AUX/IAAs to understand their underlying mechanisms of evolution. The ratios of TIR1/AFBs to AUX/IAAs range from 4:2 in Physcomitrium patens to 6:29 in Arabidopsis thaliana and 3:16 in Fragaria vesca. Whole-genome duplication (WGD) and tandem duplication have contributed to the expansion of the AUX/IAA gene family, but numerous TIR1/AFB gene duplicates were lost after WGD. We further analyzed the expression profiles of TIR1/AFBs and AUX/IAAs in different tissue parts of Physcomitrium patens, Selaginella moellendorffii, Arabidopsis thaliana and Fragaria vesca, and found that TIR1/AFBs and AUX/IAAs were highly expressed in all tissues in P. patens, S. moellendorffii. In A. thaliana and F. vesca, TIR1/AFBs maintained the same expression pattern as the ancient plants with high expression in all tissue parts, while AUX/IAAs appeared tissue-specific expression. In F. vesca, 11 AUX/IAAs interacted with TIR1/AFBs with different interaction strengths, and the functional specificity of AUX/IAAs was related to their ability to bind TIR1/AFBs, thus promoting the development of specific higher plant organs. Verification of the interactions among TIR1/AFBs and AUX/IAAs in Marchantia polymorpha and F. vesca also showed that the regulation of AUX/IAA members by TIR1/AFBs became more refined over the course of plant evolution. Conclusions Our results indicate that specific interactions and specific gene expression patterns both contributed to the functional diversification of TIR1/AFBs and AUX/IAAs.

BackgroundCircular RNAs (circRNAs) are stable, often highly expressed RNA transcripts with potential to modulate other regulatory RNAs. A few circRNAs have been shown to bind RNA-binding proteins (RBPs); however, little is known about the prevalence and distribution of these interactions in different biological contexts.MethodsWe conduct an extensive screen of circRNA-RBP interactions in the ENCODE cell lines HepG2 and K562. We profile circRNAs in deep-sequenced total RNA samples and analyze circRNA-RBP interactions using a large set of eCLIP data with binding sites of 150 RBPs. We validate interactions for select circRNAs and RBPs by performing RNA immunoprecipitation and functionally characterize our most interesting candidates by conducting knockdown studies followed by RNA-Seq.ResultsWe generate a comprehensive catalog of circRNA-RBP interactions in HepG2 and K562 cells. We show that KHSRP binding sites are enriched in flanking introns of circRNAs and that KHSRP depletion affects circRNA biogenesis. We identify circRNAs that are highly covered by RBP binding sites and experimentally validate individual circRNA-RBP interactions. We show that circCDYL, a highly expressed circRNA with clinical and functional implications in bladder cancer, is almost completely covered with GRWD1 binding sites in HepG2 cells, and that circCDYL depletion counteracts the effect of GRWD1 depletion. Furthermore, we confirm interactions between circCDYL and RBPs in bladder cancer cells and demonstrate that circCDYL depletion affects hallmarks of cancer and perturbs the expression of key cancer genes, e.g., TP53. Finally, we show that elevated levels of circCDYL are associated with overall survival of bladder cancer patients.ConclusionsOur study demonstrates transcriptome-wide and cell-type-specific circRNA-RBP interactions that could play important regulatory roles in tumorigenesis.

Abstract Background The growing application of artificial intelligence (AI) in healthcare has brought technological breakthroughs to traditional diagnosis and treatment, but it is accompanied by many risks and challenges. These adverse effects are also seen as ethical issues and affect trustworthiness in medical AI and need to be managed through identification, prognosis and monitoring. Methods We adopted a multidisciplinary approach and summarized five subjects that influence the trustworthiness of medical AI: data quality, algorithmic bias, opacity, safety and security, and responsibility attribution, and discussed these factors from the perspectives of technology, law, and healthcare stakeholders and institutions. The ethical framework of ethical values-ethical principles-ethical norms is used to propose corresponding ethical governance countermeasures for trustworthy medical AI from the ethical, legal, and regulatory aspects. Results Medical data are primarily unstructured, lacking uniform and standardized annotation, and data quality will directly affect the quality of medical AI algorithm models. Algorithmic bias can affect AI clinical predictions and exacerbate health disparities. The opacity of algorithms affects patients’ and doctors’ trust in medical AI, and algorithmic errors or security vulnerabilities can pose significant risks and harm to patients. The involvement of medical AI in clinical practices may threaten doctors ‘and patients’ autonomy and dignity. When accidents occur with medical AI, the responsibility attribution is not clear. All these factors affect people’s trust in medical AI. Conclusions In order to make medical AI trustworthy, at the ethical level, the ethical value orientation of promoting human health should first and foremost be considered as the top-level design. At the legal level, current medical AI does not have moral status and humans remain the duty bearers. At the regulatory level, strengthening data quality management, improving algorithm transparency and traceability to reduce algorithm bias, and regulating and reviewing the whole process of the AI industry to control risks are proposed. It is also necessary to encourage multiple parties to discuss and assess AI risks and social impacts, and to strengthen international cooperation and communication.

Abstract Background GRAS genes formed one of the important transcription factor gene families in plants, had been identified in several plant species. The family genes were involved in plant growth, development, and stress resistance. However, the comparative analysis of GRAS genes in Rosaceae species was insufficient. Results In this study, a total of 333 GRAS genes were identified in six Rosaceae species, including 51 in strawberry (Fragaria vesca), 78 in apple (Malus domestica), 41 in black raspberry (Rubus occidentalis), 59 in European pear (Pyrus communis), 56 in Chinese rose (Rosa chinensis), and 48 in peach (Prunus persica). Motif analysis showed the VHIID domain, SAW motif, LR I region, and PFYRE motif were considerably conserved in the six Rosaceae species. All GRAS genes were divided into 10 subgroups according to phylogenetic analysis. A total of 15 species-specific duplicated clades and 3 lineage-specific duplicated clades were identified in six Rosaceae species. Chromosomal localization presented the uneven distribution of GRAS genes in six Rosaceae species. Duplication events contributed to the expression of the GRAS genes, and Ka/Ks analysis suggested the purification selection as a major force during the evolution process in six Rosaceae species. Cis-acting elements and GO analysis revealed that most of the GRAS genes were associated with various environmental stress in six Rosaceae species. Coexpression network analysis showed the mutual regulatory relationship between GRAS and bZIP genes, suggesting the ability of the GRAS gene to regulate abiotic stress in woodland strawberry. The expression pattern elucidated the transcriptional levels of FvGRAS genes in various tissues and the drought and salt stress in woodland strawberry, which were verified by RT-qPCR analysis. Conclusions The evolution and functional analysis of GRAS genes provided insights into the further understanding of GRAS genes on the abiotic stress of Rosaceae species.

Abstract Background Salinization is major abiotic stress limiting cotton production. Melatonin (MT) has been implicated in salt stress tolerance in multiple crops including upland cotton. Here, we explored the transcriptomic and metabolomic response of a salt-tolerant self-bred high-yielding cotton line SDS-01, which was exogenously sprayed with four MT concentrations (50, 100, 200, and 500 μM). Results Here we found that MT improves plant biomass and growth under salt stress. The combined transcriptome sequencing and metabolome profiling approach revealed that photosynthetic efficiency is improved by increasing the expressions of chlorophyll metabolism and antenna proteins in MT-treated seedlings. Additionally, linoleic acid and flavonoid biosynthesis were improved after MT treatment. The Na+/K+ homeostasis-related genes were increasingly expressed in salt-stressed seedlings treated with MT as compared to the ones experiencing only salt stress. Melatonin treatment activated a cascade of plant-hormone signal transduction and reactive oxygen scavenging genes to alleviate the detrimental effects of salt stress. The global metabolome profile revealed an increased accumulation of flavonoids, organic acids, amino acids and derivatives, saccharides, and phenolic acids in MT-treated seedlings. Interestingly, N, N′-Diferuloylputrescine a known antioxidative compound was highly accumulated after MT treatment. Conclusion Collectively, our study concludes that MT is a salt stress regulator in upland cotton and alleviates salt-stress effects by modulating the expressions of photosynthesis (and related pathways), flavonoid, ROS scavenging, hormone signaling, linoleic acid metabolism, and ion homeostasis-related genes.