Ask about this productRelated genes to: GPT2 Blocking Peptide
- Gene:
- GPT2 NIH gene
- Name:
- glutamic--pyruvic transaminase 2
- Previous symbol:
- -
- Synonyms:
- ALT2
- Chromosome:
- 16q11.2
- Locus Type:
- gene with protein product
- Date approved:
- 2002-03-05
- Date modifiied:
- 2016-10-05
Related products to: GPT2 Blocking Peptide
Related articles to: GPT2 Blocking Peptide
- This paper proposes a parameter-efficient fine-tuning framework that integrates uncertainty modeling with external memory augmentation, aiming to improve robustness, confidence calibration, and contextual completeness in downstream natural language processing tasks. From the methodological perspective, the uncertainty modeling module explicitly characterizes uncertainty in inputs and intermediate representations through feature-level estimation, cross-layer propagation, and confidence calibration, thereby enhancing training stability and reducing the influence of noisy signals. Meanwhile, the external memory augmentation module employs key-value retrieval and gated fusion mechanisms to provide reusable contextual support, alleviating information loss caused by limited contextual summarization and improving representation quality under heterogeneous evaluation settings. Extensive experiments and ablation studies were conducted on text classification and named entity recognition tasks across multiple public benchmark datasets, using GPT-2 Small, GPT-2 Medium, and LLaMA3-8B as backbone models. The results demonstrate that the proposed framework consistently outperforms several mainstream fine-tuning methods in terms of accuracy, F1 score, and robustness, while also showing stable behavior under learning-rate sensitivity and missing-information settings. Overall, this study provides a novel perspective for efficient and interpretable fine-tuning paradigms, achieving a favorable balance among performance improvement, parameter efficiency, and deployment feasibility, and offering a practical basis for future extensions to more complex downstream scenarios. - Source: PubMed
Publication date: 2026/06/12
Ma YumengXing YueWu DiZhou YiningZi YunWang MingDeng YingnanPan Shuaidong - Craniosynostosis is defined by premature cranial suture fusion and is biologically heterogeneous. To map mitochondrial-associated signals in craniosynostosis and rank follow-up candidates, we integrated two public microarray datasets (GSE27976, GSE50796), corrected batch effects, and analyzed 14,186 shared genes using limma. This identified 798 nominal DEGs (388 upregulated and 410 downregulated), of which 19 remained significant after Benjamini-Hochberg correction. Intersecting the nominal DEG list with the MitoCarta 3.0 inventory yielded 24 mitochondrial DEGs (MitoDEGs). Complementary feature selection reduced these 24 MitoDEGs to an eight-gene panel (TMEM11, SLC25A21, GPT2, CYP27A1, MRPS30, ACAA2, GSR, and LIG3); a multigene score reached an apparent AUC of 0.806 in the integrated dataset. Correlation-based co-expression analyses linked the panel to mitochondrial translation, electron transport, amino-acid metabolism, redox control, and cell-matrix signaling. Among craniosynostosis cases, consensus clustering on the eight genes separated two molecular subtypes with distinct GSVA pathway profiles. For experimental support in a genetically defined mouse model, we profiled bilateral coronal suture complexes from Fgfr2C361Y/+ knock-in (KI) pups and WT littermates. Jess capillary immunoassay showed higher CYP27A1 abundance in KI sutures (P = 0.0276), whereas ACAA2, LIG3, MRPS30, and TMEM11 were not significant. Data-independent acquisition (DIA) proteomics identified 523 differentially abundant proteins (516 increased, 7 decreased in KI), followed by stricter-threshold reporting, sensitivity analysis, and threshold-free rank-based enrichment. MitoCarta proteins and mitochondrial pathways remained supported under these more conservative analyses. These results support mitochondria-associated transcriptomic and proteomic changes in craniosynostosis and prioritize a limited set of mitochondrial candidates for future mechanistic work. - Source: PubMed
Publication date: 2026/06/08
Zeng HanWang YuDong MiaoYue YingyingJin Xiaolei - This study proposes a multimodal and AI-supported approach for both generation and detection of vishing threats, also known as audio social engineering attacks. By considering text and audio modalities together, fake contents specific to Turkish language were generated with GPT-2 and Tacotron-2 models; detection processes were carried out with discriminator structures such as BERT, CNN, LSTM, and Whisper. The developed system was designed with an integrated analysis (multimodal fusion) approach at feature and decision levels and was evaluated at both technical and psychosocial levels. In user tests conducted with 394 participants, the rate of perception of fake contents as real was measured in the range of 92.8-95.3%; it was observed that contents containing official institution imitations were the most convincing class. Experimental results show that the F1 scores obtained with single models remained in the range of 0.86-0.91%; however, 96.8% accuracy and 0.947 F1 score were achieved with the BERT + CNN + Whisper architecture. It has also been determined that GPT-4-supported content generation increases the plausibility of attack scenarios and challenges the performance of classifier models. The study is a pioneer in the field of fake audio content generation and detection in Turkish and reveals how LLM-supported multimodal systems can be used in both individual and institutional defense applications. - Source: PubMed
Publication date: 2026/05/23
Albayrak Ahmet - As the most aggressive form of breast cancer, triple-negative breast cancer (TNBC) is associated with poor prognosis and a lack of effective therapeutic options. Glycosylation has been linked to metabolic reprogramming in various cancers, and therapies targeting glycosylation-mediated metabolic reprogramming have been found to be effective. However, the role of the glycosyltransferase GALNT6 in TNBC metabolic reprogramming has not been examined. - Source: PubMed
Publication date: 2026/05/21
Bian ZimingLiu YangGuo ShiqunHu DongxueLiu DanhuiFan SairongChen Xiaoming - To evaluate and compare the performance of large language models (LLMs) in identifying contributing factors (CFs) underlying patient safety incident investigations. - Source: PubMed
Publication date: 2026/05/20
Wang YingBowditch LorelleMolloy CharlotteYu YinghuaHibbert PeterMagrabi Farah