Objectives: Colorectal cancer (CRC) is a common carcinoma of the gastrointestinal tract with high incidence and mortality worldwide. Studies have shown that long noncoding RNAs (lncRNAs) play important roles in CRC. Our purpose is to investigate the potential of serum Linc01836 as a diagnostic and prognostic marker in CRC. Methods: We evaluated the expression of Linc01836 via quantitative real-time polymerase chain reaction (qRT-PCR). The serum CEA, CA19-9, Cyfra21-1, and CA72-4 concentrations were measured by Architect I4000 SR. Receiver operating characteristic (ROC) curves were plotted to estimate the diagnostic value in CRC. Relationship between serum Linc01836 expression and clinicopathological characteristics of CRC cases was analyzed via chi-square test. The underlying mechanism of Linc01836 on the development and prognosis in CRC was predicted by bioinformatic analysis. Results: The method of qRT-PCR for Linc01836 detection was confirmed with high precision and specificity. Serum Linc01836 expression in CRC patients was significantly higher than that in healthy donors (p < 0.0001) and benign patients (p < 0.0001), and declined after resection (p < 0.01). High expression of Linc01836 was associated with histological stage (p = 0.002) and lymph node metastasis (p = 0.006). In addition, serum Linc01836 could effectively differentiate CRC patients from the healthy folks, with favorable area under the curve (AUC) of 0.809 (95% CI: 0.757-0.861, p < 0.001). What is more, the combination of serum Linc01836, CEA, and Cyfra21-1 could improve diagnostic sensitivity (92.0%). Linc01836 was averagely located in the nucleus and cytoplasm, suggesting that it might participate in CRC progression and prognosis through the crosstalk among lncRNAs, miRNAs, and mRNAs. Conclusion: Linc01836 may serve as a valuable noninvasive biomarker for population screening, early detection, and clinical surveillance of CRC.
Motivation: We present a sequence-based framework and algorithm PHYLOCLUS for predicting co-regulated genes. In our approach, de novo discovery methods are used to find motifs conserved by evolution and then a Bayesian hierarchical clustering model is used to cluster these motifs, thereby grouping together genes that are putatively co-regulated. Our clustering procedure allows both the number of clusters and the motif width within each cluster to be unknown.
Based on interim analyses and modeling data, lower doses of bamlanivimab and etesevimab together (700/1400 mg) were investigated to determine optimal dose and expand availability of treatment.This Phase 3 portion of the BLAZE-1 trial characterized the effect of bamlanivimab with etesevimab on overall patient clinical status and virologic outcomes in ambulatory patients ≥12 years old, with mild-to-moderate coronavirus disease 2019 (COVID-19), and ≥1 risk factor for progressing to severe COVID-19 and/or hospitalization. Bamlanivimab and etesevimab together (700/1400 mg) or placebo were infused intravenously within 3 days of patients' first positive COVID-19 test.In total, 769 patients were infused (median age [range]; 56.0 years [12, 93], 30.3% of patients ≥65 years of age and median duration of symptoms; 4 days). By day 29, 4/511 patients (0.8%) in the antibody treatment group had a COVID-19-related hospitalization or any-cause death, as compared with 15/258 patients (5.8%) in the placebo group (Δ[95% confidence interval {CI}] = -5.0 [-8.0, -2.1], P < .001). No deaths occurred in the bamlanivimab and etesevimab group compared with 4 deaths (all COVID-19-related) in the placebo group. Patients receiving antibody treatment had a greater mean reduction in viral load from baseline to Day 7 (Δ[95% CI] = -0.99 [-1.33, -.66], P < .0001) compared with those receiving placebo. Persistently high viral load at Day 7 correlated with COVID-19-related hospitalization or any-cause death by Day 29 in all BLAZE-1 cohorts investigated.These data support the use of bamlanivimab and etesevimab (700/1400 mg) for ambulatory patients at high risk for severe COVID-19. Evolution of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants will require continued monitoring to determine the applicability of this treatment.NCT04427501.
Recent developments in deep learning have led to great success in various natural language processing (NLP) tasks. However, these applications may involve data that contain sensitive information. Therefore, how to achieve good performance while also protecting the privacy of sensitive data is a crucial challenge in NLP. To preserve privacy, Differential Privacy (DP), which can prevent reconstruction attacks and protect against potential side knowledge, is becoming a de facto technique for private data analysis. In recent years, NLP in DP models (DP-NLP) has been studied from different perspectives, which deserves a comprehensive review. In this paper, we provide the first systematic review of recent advances in DP deep learning models in NLP. In particular, we first discuss some differences and additional challenges of DP-NLP compared with the standard DP deep learning. Then, we investigate some existing work on DP-NLP and present its recent developments from three aspects: gradient perturbation based methods, embedding vector perturbation based methods, and ensemble model based methods. We also discuss some challenges and future directions.
Little is known about the effects of chronic alcohol intake on the outcome of acute kidney injury (AKI). Hence, we examined the effects of chronic alcohol intake on the development of renal fibrosis following AKI in an animal model of bilateral renal ischemia-reperfusion (IR) injury. We first found that chronic alcohol exposure exacerbated bilateral IR-induced renal fibrosis and renal function impairment. This phenomenon was associated with increased bilateral IR-induced extracellular matrix deposition and an increased myofibroblast population as well as increased bilateral IR-induced expression of fibrosis-related genes in the kidneys. To explore the mechanisms underlying this phenomenon, we showed that chronic alcohol exposure enhanced β-arrestin 2 (Arrb2) expression and Akt and glycogen synthase kinase-3 (GSK3)β activation in the kidneys. Importantly, pharmacological GSK3 inhibition alleviated bilateral IR-induced renal fibrosis and renal function impairment. Furthermore, we demonstrated that Arrb2-/- mice exhibited resistance to IR-induced renal fibrosis and renal function impairment following chronic alcohol exposure, and these effects were associated with attenuated GSK3β activation in the kidneys. Taken together, our results suggest that chronic alcohol exposure may potentiate AKI via β-arrestin 2/Akt/GSK3β-mediated signaling in the kidney.
Recently developed large language models (LLMs) such as ChatGPT, Claude, and Llama have demonstrated impressive abilities, and even surpass human-level performance in several tasks. Despite their success, the resource-intensive demands of these models, requiring significant computational power for both training and inference, limit their deployment to high-performance servers. Additionally, the extensive calculation requirements of the models often lead to increased latency in response times. With the increasing need for LLMs to operate efficiently on CPUs, research about lightweight models that are optimized for CPU inference has emerged. In this work, we introduce GEB-1.3B, a lightweight LLM trained on 550 billion tokens in both Chinese and English languages. We employ novel training techniques, including ROPE, Group-Query-Attention, and FlashAttention-2, to accelerate training while maintaining model performance. Additionally, we fine-tune the model using 10 million samples of instruction data to enhance alignment. GEB-1.3B exhibits outstanding performance on general benchmarks such as MMLU, C-Eval, and CMMLU, outperforming comparative models such as MindLLM-1.3B and TinyLLaMA-1.1B. Notably, the FP32 version of GEB-1.3B achieves commendable inference times on CPUs, with ongoing efforts to further enhance speed through advanced quantization techniques. The release of GEB-1.3B as an open-source model marks a significant contribution to the development of lightweight LLMs, promising to foster further research and innovation in the field.
Background : Prasugrel and clopidogrel require transformation into active metabolites by cytochrome P450 (CYP) enzymes. Variants in CYP genes, particularly CYP2C19 , are associated w/ reduced levels of metabolite and platelet inhibition w/ clopidogrel but not prasugrel. We investigated the impact of variants in CYP genes on clinical outcomes. Methods : TRITON-TIMI 38 randomized ACS patients w/ planned PCI to prasugrel (60 mg load, 10 mg qd) or clopidogrel (300 mg, 75 mg qd) for 6–15 mos. DNA was available in 2943 subjects who were genotyped for functional polymorphisms in CYP2C19 , 2C9 , 3A4/5 , 2B6 , and 1A2 and classified for each gene as carriers or non-carriers of ≥1 reduced function allele. Results : Among subjects treated with clopidogrel, CYP2C19 reduced function allele carriers (27% of population) had a 53% increased risk of the primary efficacy endpoint of CV death, MI, or stroke (HR 1.53, P=0.007, Fig ) and a 3-fold increased risk of ARC-defined definite or probable stent thrombosis (HR 3.09, P=0.038) compared w/ non-carriers. In contrast, with prasugrel CYP2C19 reduced function allele carriers were not at increased risk of CV death, MI, or stroke (HR 0.89, P=0.86) or stent thrombosis (HR 0.60, P=0.57). No association between genotype and bleeding was observed w/ either treatment. There were no statistically significant associations between any other tested CYP genotypes and outcomes. Conclusions : CYP2C19 genetic variants identify patients at significantly higher risk for CV death, MI, or stroke and for stent thrombosis among those treated with clopidogrel but not among those treated with prasugrel. CYP2C19 genotyping offers the potential to help tailor antiplatelet therapy.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
