To the Editor, We would like to thank the rigorous dataset inspection work conducted by the author of “Addressing flaws in the Seq2Topt dataset for the prediction of enzyme optimal temperature”. The dataset of enzyme optimal temperature (Topt) used to develop Seq2Topt [1] was obtained from https://github.com/jafetgado/tomer, which was curated by Li et al. 2019 [2] and Gado et al. 2020 [3] from the BRENDA database [4]. In “Addressing flaws in the Seq2Topt dataset for the prediction of enzyme optimal temperature,” the author stated that the Topt values of F9FU71, A9U908, J7HET3, P94368, Q5YXQ1, Q83V33, Q50539, Q50538, Q2WCS9 are erroneously annotated in the dataset used by Seq2Topt. Here, we presented all references associated with these data entries in the BRENDA database in Table 1, and the Topt values of F9FU71, P94368, Q5YXQ1, Q83V33, Q50539, and Q50538 were not found in relevant references [5–9]. Dhar et al. 2013 [10] stated that the Topt values of A9U908 (sodA) and J7HET3 (sodB) were 4°C, but we agree with the critique that no enzyme activity measurements at lower temperatures were conducted to confirm that 4°C was the Topt value of A9U908 and J7HET3. For Q2WCS9, De Angelis et al. 2010 [11] reported that 13°C was the measured Topt value, whereas the critique letter did not provide a source for the claimed measured Topt value of 37°C. Selected nine entries from Seq2Topt dataset. Overall, we concur with the critique that the dataset used to develop enzyme Topt predictive models (and machine learning models for other protein properties) should be carefully reviewed and revised to ensure validity, rather than simply extracting data from established databases such as BRENDA. While such curation may be costly, it remains necessary until a high level of validity is automatically guaranteed by these biochemical databases. Accordingly, future updates of Seq2Topt will be based on a more rigorously vetted dataset. The author declares that there is no conflict of interest. None declared. No new data was generated in this study. This response letter refers to the data provided in the Seq2Topt study [1].

An accurate deep learning predictor is needed for enzyme optimal temperature (${T}_{opt}$), which quantitatively describes how temperature affects the enzyme catalytic activity. In comparison with existing models, a new model developed in this study, Seq2Topt, reached a superior accuracy on ${T}_{opt}$ prediction just using protein sequences (RMSE = 12.26°C and R2 = 0.57), and could capture key protein regions for enzyme ${T}_{opt}$ with multi-head attention on residues. Through case studies on thermophilic enzyme selection and predicting enzyme ${T}_{opt}$ shifts caused by point mutations, Seq2Topt was demonstrated as a promising computational tool for enzyme mining and in-silico enzyme design. Additionally, accurate deep learning predictors of enzyme optimal pH (Seq2pHopt, RMSE = 0.88 and R2 = 0.42) and melting temperature (Seq2Tm, RMSE = 7.57 °C and R2 = 0.64) were developed based on the model architecture of Seq2Topt, suggesting that the development of Seq2Topt could potentially give rise to a useful prediction platform of enzymes.

The enzyme turnover rate, ${k}_{cat}$, quantifies enzyme kinetics by indicating the maximum efficiency of enzyme catalysis. Despite its importance, ${k}_{cat}$ values remain scarce in databases for most organisms, primarily because of the cost of experimental measurements. To predict ${k}_{cat}$ and account for its strong temperature dependence, DLTKcat was developed in this study and demonstrated superior performance (log10-scale root mean squared error = 0.88, R-squared = 0.66) than previously published models. Through two case studies, DLTKcat showed its ability to predict the effects of protein sequence mutations and temperature changes on ${k}_{cat}$ values. Although its quantitative accuracy is not high enough yet to model the responses of cellular metabolism to temperature changes, DLTKcat has the potential to eventually become a computational tool to describe the temperature dependence of biological systems.

Beamforming plays an important role in 5G Massive Multiple-Input Multiple-Output (MMIMO) communications. Optimizing beamforming configurations for 5G base stations (BSs) can substantially improve the quality of service for mobile users, and thus has great practical value. However, identifying the optimal beamforming configurations has proven to be a complex task, and is even more challenging when accounting for the unavoidable coupled influence among multiple densely deployed 5G BSs. In this paper, we propose a highly efficient deep multi-agent Bayesian optimization framework for the coordinated beamforming optimization problem that involves multiple BSs. Its core algorithm is built upon a deep ensemble of neural networks and a sample-efficient upper confidence bound (UCB) based exploration strategy. Our numerical results show that the proposed approach is highly effective in searching for optimally coordinated beamforming vectors under extremely large search space, and beats strong multi-agent reinforcement learning baselines in terms of optimization quality and sample efficiency.

In microorganisms, different from primary metabolism for cellular growth, secondary metabolism is for ecological interactions and stress responses and an important source of natural products widely used in various areas such as pharmaceutics and food additives. With advancements of sequencing technologies and bioinformatics tools, a large number of biosynthetic gene clusters of secondary metabolites have been discovered from microbial genomes. However, due to challenges from the difficulty of genome-scale pathway reconstruction and the limitation of conventional flux balance analysis (FBA) on secondary metabolism, the quantitative modeling of secondary metabolism is poorly established, in contrast to that of primary metabolism. This review first discusses current efforts on the reconstruction of secondary metabolic pathways in genome-scale metabolic models (GSMMs), as well as related FBA-based modeling techniques. Additionally, potential extensions of FBA are suggested to improve the prediction accuracy of secondary metabolite production. As this review posits, biosynthetic pathway reconstruction for various secondary metabolites will become automated and a modeling framework capturing secondary metabolism onset will enhance the predictive power. Expectedly, an improved FBA-based modeling workflow will facilitate quantitative study of secondary metabolism and in silico design of engineering strategies for natural product production.

Data security policies have significant impacts on big data and artificial intelligence applications, and yield data silo due to the data inaccessibility among the commercial companies for privacy-preserving. On September 1st, 2021, China officially implemented the Data Security Law. This paper timely proposes, validates, and productizes a trusted data sharing solution based on Vertical Federated Learning (VFL) technology across telecom and finance companies. A VFL model with Hetero Secure Boost Tree (HSBT) algorithm is proposed to overcome the data silo problems and to preserve user data privacy. Experimental results demonstrate that the model with both financial and telecom data improves the success rate of marketing for a tier-1 commercial bank in China. The solution has also been successfully implemented to greatly improve the marketing efficiency, resulting an approximately 50% cost reduction.

BACKGROUND: The rising consensus that the cell can dynamically allocate its resources provides an interesting angle for discovering the governing principles of cell growth and metabolism. Extensive efforts have been made in the past decade to elucidate the relationship between resource allocation and phenotypic patterns of microorganisms. Despite these exciting developments, there is still a lack of explicit comparison between potentially competing propositions and a lack of synthesis of inter-related proposals and findings. RESULTS: In this work, we have reviewed resource allocation-derived principles, hypotheses and mathematical models to recapitulate important achievements in this area. In particular, the emergence of resource allocation phenomena is deciphered by the putative tug of war between the cellular objectives, demands and the supply capability. Competing hypotheses for explaining the most-studied phenomenon arising from resource allocation, i.e. the overflow metabolism, have been re-examined towards uncovering the potential physiological root cause. The possible link between proteome fractions and the partition of the ribosomal machinery has been analysed through mathematical derivations. Finally, open questions are highlighted and an outlook on the practical applications is provided. It is the authors' intention that this review contributes to a clearer understanding of the role of resource allocation in resolving bacterial growth strategies, one of the central questions in microbiology. CONCLUSIONS: We have shown the importance of resource allocation in understanding various aspects of cellular systems. Several important questions such as the physiological root cause of overflow metabolism and the correct interpretation of 'protein costs' are shown to remain open. As the understanding of the mechanisms and utility of resource application in cellular systems further develops, we anticipate that mathematical modelling tools incorporating resource allocation will facilitate the circuit-host design in synthetic biology.

Mobile number portability (MNP) policy highlights the problem ofsubscriber churn and thus enhance the liquidity and competition of the telecommunication market. China will implement MNP policy on November $1^{\mathrm{s}\mathrm{t}}$, 2019 after 9 years of trials. This paper timely proposes, validates, and productizes a game theory assisted machine learning scheme to help the mobile network operator (MNO) in China make a strategy to proactively cope with their competitors in the same MNP market. The scheme further develops a set of MNP tactics for the MNO to detect user churn behaviors and to remedy the users with appropriate treatments. Experimental results demonstrate that the scheme can guide the MNOs to make targeted MNP strategy and precisely identify the “abnormal” subscribers who tend to churn out and potential new subscribers who may churn in. The scheme has been successfully implemented in production to greatly improve the marketing efficiency and user satisfaction in terms of an approximately 50% reduction of user churn for a tier-1 MNO in China.

A variance detection (VD)-based impulse radio ultra-wideband (IR-UWB) system has the potential of mitigating narrowband interference (NBI), even in multipath environments. However, as a highly nonlinear system, which involves fourth-order multiplications, comprehensive analysis and evaluations of its performances have not been seen so far. In this paper, we address the issues by developing a log-normal random distribution model for the VD-based IR-UWB receiver and deriving the analytical bit-error-rate formulas for the system in both additive white Gaussian noise and CM1 multipath channels with and without NBIs for the first time. Furthermore, through both theoretical analysis and simulations, we show that the presented VD-based receivers can outperform the conventional energy-detection-based and the fourth-order detection-based receivers and have a stronger inherent ability to mitigate destructive performance degradation caused by strong NBIs. This makes the VD-based IR-UWB system a good candidate as a non-coherent IR-UWB receiver for low power and low complexity applications.

In this paper, a unified framework, denoted as generalized nonlinear detection (GND), is developed to encompass existing nonlinear detection technologies as well as further optimize the performance of nonlinear detection for noncoherent impulse radio (IR) ultra-wideband (UWB) receivers. Computer simulations show that independent of absence or presence of narrowband interferences (NBIs), the GND-based receivers can achieve much better bit-error rate (BER) performance than the energy detection (ED) based receivers under both AWGN and multipath channels. Furthermore, as compared to the existing nonlinear detection algorithms, the GND algorithm with a well selected β value always has the best BER performance and the strongest ability to resist NBIs.

Multiscale modelling as an emerging modelling paradigm is now widely regarded as a promising and powerful tool in various disciplines. However, a multiscale model is usually much more difficult to develop than a single-scale model due to a range of challenges. This work presents a methodology to facilitate the development of multiscale models, which comprises three main modelling steps, namely conceptual modelling, model realization and model execution. A set of proof-of-concept tools have been developed to realize the proposed methodology. A case study on the modelling of a heterogeneous chemical reactor is presented to demonstrate these tools and to illustrate the key concepts. © ECMS.