Group 1 - The AI model AlphaCD has been developed by the Agricultural Genome Research Institute of the Chinese Academy of Agricultural Sciences, which can efficiently predict the enzymatic activity characteristics of over 20,000 cytosine deaminases and design new high-performance base editing tools [2] - The Central Education Work Leading Group has issued an action plan for optimizing higher education discipline and major settings from 2025 to 2027, focusing on strategic emerging industries and future industries [2] - The State Administration for Market Regulation has released a list of food adulteration testing methods, addressing three major technical challenges in food safety [3] Group 2 - Nanyang has introduced new housing policies to promote the healthy development of the real estate market in the central urban area, effective from August 13, 2023, until December 31, 2026, with 20 measures aimed at supporting housing demand and improving market confidence [4] - UK researchers have achieved a breakthrough by chemically connecting RNA with amino acids, providing new insights into the origins of life [5] - Google announced an additional investment of $9 billion in Virginia for cloud computing and AI infrastructure by 2026, part of a broader commitment to AI training for all university students in the state [6] Group 3 - NVIDIA reported Q2 revenue of $46.7 billion for fiscal year 2026, up from $30.04 billion in the same period last year, exceeding market expectations [7] - Meituan's Q2 revenue reached 91.8 billion yuan, a year-on-year increase of 11.7%, but the company anticipates significant losses in its core local business for Q3 due to intense competition in the food delivery market [11] - Japanese property insurance company SOMPO has agreed to acquire insurance company Aspen for approximately $3.5 billion [10]

Group 1 - OpenAI has released GPT-Realtime, a multimodal model designed for voice AI agents, capable of generating natural and fluent speech, mimicking human tones, emotions, and speech rates, suitable for various sectors including customer service, education, finance, and healthcare [2] - The Chinese Academy of Agricultural Sciences has developed AlphaCD, an AI model that predicts enzyme activity characteristics and designs new high-performance base editing tools, based on the largest experimental validation dataset globally [2] - Alibaba's Lingyang has launched a data analysis agent, upgrading its Quick BI's "Smart Q" with three core capabilities: querying, interpretation, and reporting, set to be fully available to enterprise users by September 9 [2] - Tencent has open-sourced its end-to-end video sound effect generation model, HunyuanVideo-Foley, allowing users to generate high-quality sound effects from video and text descriptions [2] Group 2 - The article mentions the availability of various investment and industry insights, including humanoid robots, commercial aerospace, and AGI, encouraging readers to join the membership for deeper analysis [3]

Core Insights - The Chinese Academy of Agricultural Sciences has developed the world's largest experimental validation dataset for agricultural gene editing technology and created an AI model named AlphaCD, which can efficiently predict the enzymatic characteristics of over 20,000 cytosine deaminase variants and design new high-performance base editing tools [1] Group 1 - The research team spent three years testing 1,100 variants of cytosine deaminase for catalytic efficiency and off-target effects, establishing a large-scale experimental database covering multidimensional functional characteristics [1] - AlphaCD significantly outperforms traditional methods in predicting key indicators such as catalytic efficiency, off-target activity, targeting window, and motif preference [1] - The development of AlphaCD addresses the limitations of traditional cytosine base editors, which rely on extensive trial-and-error experiments that are costly and time-consuming [1] Group 2 - AlphaCD overcomes the limitations of current AI models that depend on supercomputing resources, allowing model training to be completed on personal computers using a lightweight design [2] - This lightweight design enables academic institutions to conduct high-throughput functional predictions at a low cost [2]

Core Viewpoint - The rapid development of genomics presents unprecedented opportunities and challenges for characterizing protein functions, highlighting the limitations of traditional biochemical methods and the need for high-precision, quantitative approaches [3][4]. Group 1: Research Development - A machine learning model named AlphaCD was developed to accurately characterize 21,335 cytidine deaminases (CD), demonstrating high precision in predicting catalytic efficiency, off-target activity, target site window, and catalytic motifs [4][6]. - The research team experimentally characterized 1,100 APOBEC-like family cytidine deaminases fused with nCas9 in HEK293T cells, generating the largest dataset of experimental functional data for a single protein family to date [6][8]. Group 2: Model Performance - AlphaCD achieved high precision in predicting catalytic efficiency (0.92), off-target activity (0.84), target site window (0.73), and catalytic motifs (0.78) [6]. - The model was validated by subsampling 28 cytidine deaminases from the Uniprot database, with prediction accuracies of 0.84, 0.87, 0.75, and 0.73 for the respective features [6][8]. Group 3: Practical Application - The research team optimized the off-target site of one cytidine deaminase (A0A2R2Z4E4) using alanine scanning mutagenesis, creating a variant (A0A2R2Z4E4 E100A) that serves as a highly accurate and efficient cytidine base editor [8]. - This case exemplifies the application value of AlphaCD in high-precision, high-throughput protein function characterization and provides a paradigm for accelerating functional analysis of other proteins [8].