Existing approaches to automatic data transformation are insufficient to meet the requirements in many real-world scenarios, such as the building sector. First, there is no convenient interface for domain experts to provide domain knowledge easily. Second, they require significant training data collection overheads. Third, the accuracy suffers from complicated schema changes. To address these shortcomings, we present a novel approach that leverages the unique capabilities of large language models (LLMs) in coding, complex reasoning, and zero-shot learning to generate SQL code that transforms the source datasets into the target datasets. We demonstrate the viability of this approach by designing an LLM-based framework, termed SQLMorpher, which comprises a prompt generator that integrates the initial prompt with optional domain knowledge and historical patterns in external databases. It also implements an iterative prompt optimization mechanism that automatically improves the prompt based on flaw detection. The key contributions of this work include (1) pioneering an end-to-end LLM-based solution for data transformation, (2) developing a benchmark dataset of 105 real-world building energy data transformation problems, and (3) conducting an extensive empirical evaluation where our approach achieved 96% accuracy in all 105 problems. SQLMorpher demonstrates the effectiveness of utilizing LLMs in complex, domain-specific challenges, highlighting the potential of their potential to drive sustainable solutions.

Background

This study aims to construct and verify a nomogram model for microvascular invasion (MVI) based on hepatocellular carcinoma (HCC) tumor characteristics and differential protein expressions, and explore the clinical application value of the prediction model.

Methods

The clinicopathological data of 200 HCC patients were collected and randomly divided into training set and validation set according to the ratio of 7:3. The correlation between MVI occurrence and primary disease, age, gender, tumor size, tumor stage, and immunohistochemical characteristics of 13 proteins, including GPC3, CK19 and vimentin, were statistically analyzed. Univariate and multivariate analyzes identified risk factors and independent risk factors, respectively. A nomogram model that can be used to predict the presence of MVI was subsequently constructed. Then, receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) were conducted to assess the performance of the model.

Results

Multivariate logistic regression analysis indicated that tumor size, GPC3, P53, RRM1, BRCA1, and ARG were independent risk factors for MVI. A nomogram was constructed based on the above six predictors. ROC curve, calibration, and DCA analysis demonstrated the good performance and the clinical application potential of the nomogram model.

Conclusions

The predictive model constructed based on the clinical characteristics of HCC tumors and differential protein expression patterns could be helpful to improve the accuracy of MVI diagnosis in HCC patients.

There is an urgent need for animal models to study SARS-CoV-2 pathogenicity. Here, we generate and characterize a novel mouse-adapted SARS-CoV-2 strain, MASCp36, that causes severe respiratory symptoms, and mortality. Our model exhibits age- and gender-related mortality akin to severe COVID-19. Deep sequencing identified three amino acid substitutions, N501Y, Q493H, and K417N, at the receptor binding domain (RBD) of MASCp36, during in vivo passaging. All three RBD mutations significantly enhance binding affinity to its endogenous receptor, ACE2. Cryo-electron microscopy analysis of human ACE2 (hACE2), or mouse ACE2 (mACE2), in complex with the RBD of MASCp36, at 3.1 to 3.7 Å resolution, reveals the molecular basis for the receptor-binding switch. N501Y and Q493H enhance the binding affinity to hACE2, whereas triple mutations at N501Y/Q493H/K417N decrease affinity and reduce infectivity of MASCp36. Our study provides a platform for studying SARS-CoV-2 pathogenesis, and unveils the molecular mechanism for its rapid adaptation and evolution.

Accurate lesion segmentation is critical in stroke rehabilitation research for the quantification of lesion burden and accurate image processing. Current automated lesion segmentation methods for T1-weighted (T1w) MRIs, commonly used in stroke research, lack accuracy and reliability. Manual segmentation remains the gold standard, but it is time-consuming, subjective, and requires neuroanatomical expertise. We previously released an open-source dataset of stroke T1w MRIs and manually-segmented lesion masks (ATLAS v1.2, N = 304) to encourage the development of better algorithms. However, many methods developed with ATLAS v1.2 report low accuracy, are not publicly accessible or are improperly validated, limiting their utility to the field. Here we present ATLAS v2.0 (N = 1271), a larger dataset of T1w MRIs and manually segmented lesion masks that includes training (n = 655), test (hidden masks, n = 300), and generalizability (hidden MRIs and masks, n = 316) datasets. Algorithm development using this larger sample should lead to more robust solutions; the hidden datasets allow for unbiased performance evaluation via segmentation challenges. We anticipate that ATLAS v2.0 will lead to improved algorithms, facilitating large-scale stroke research.

Since the last IAEA-FEC in 2021, significant progress on the development of long pulse steady state scenario and its related key physics and technologies have been achieved, including the reproducible 403 s long-pulse steady-state H-mode plasma with pure radio frequency (RF) power heating. A thousand-second time scale (∼1056 s) fully non-inductive plasma with high injected energy up to 1.73 GJ has also been achieved. The EAST operational regime of high βP has been significantly extended (H98y2 > 1.3, βP ∼ 4.0, βN ∼ 2.4 and ne/nGW ∼ 1.0) using RF and neutral beam injection (NBI). The full edge localized mode suppression using the n = 4 resonant magnetic perturbations has been achieved in ITER-like standard type-I ELMy H-mode plasmas with q95 ≈ 3.1 on EAST, extrapolating favorably to the ITER baseline scenario. The sustained large ELM control and stable partial detachment have been achieved with Ne seeding. The underlying physics of plasma-beta effect for error field penetration, where toroidal effect dominates, is disclosed by comparing the results in cylindrical theory and MARS-Q simulation in EAST. Breakdown and plasma initiation at low toroidal electric fields (<0.3 V m−1) with EC pre-ionization is developed. A beneficial role on the lower hybrid wave injection to control the tungsten concentration in the NBI discharge is observed for the first time in EAST suggesting a potential way toward steady-state H-mode NBI operation.