PhosAF: An integrated deep learning architecture for predicting protein phosphorylation sites with AlphaFold2 predicted structures.

Phosphorylation is indispensable in comprehending biological processes, while biological experimental methods for identifying phosphorylation sites are tedious and arduous. With the rapid growth of biotechnology, deep learning methods have made significant progress in site prediction tasks. Nevertheless, most existing predictors only consider protein sequence information, that limits the capture of protein spatial information. Building upon the latest advancement in protein structure prediction by AlphaFold2, a novel integrated deep learning architecture PhosAF is developed to predict phosphorylation sites in human proteins by integrating CMA-Net and MFC-Net, which considers sequence and structure information predicted by AlphaFold2. Here, CMA-Net module is composed of multiple convolutional neural network layers and multi-head attention is appended to obtaining the local and long-term dependencies of sequence features. Meanwhile, the MFC-Net module composed of deep neural network layers is used to capture the complex representations of evolutionary and structure features. Furthermore, different features are combined to predict the final phosphorylation sites. In addition, we put forward a new strategy to construct reliable negative samples via protein secondary structures. Experimental results on independent test data and case study indicate that our model PhosAF surpasses the current most advanced methods in phosphorylation site prediction.

STAGAN: An approach for improve the stability of molecular graph generation based on generative adversarial networks.

With the wide application of deep learning in Drug Discovery, deep generative model has shown its advantages in drug molecular generation. Generative adversarial networks can be used to learn the internal structure of molecules, but the training process may be unstable, such as gradient disappearance and model collapse, which may lead to the generation of molecules that do not conform to chemical rules or a single style. In this paper, a novel method called STAGAN was proposed to solve the difficulty of model training, by adding a new gradient penalty term in the discriminator and designing a parallel layer of batch normalization used in generator. As an illustration of method, STAGAN generated higher valid and unique molecules than previous models in training datasets from QM9 and ZINC-250K. This indicates that the proposed method can effectively solve the instability problem in the model training process, and can provide more instructive guidance for the further study of molecular graph generation.

Generation of focused drug molecule library using recurrent neural network.

CONTEXT: With the wide application of deep learning in drug research and development, de novo molecular design methods based on recurrent neural network (RNN) have strong advantages in drug molecule generation. The RNN model can be used to learn the internal chemical structure of molecules, which is similar to a natural language processing task. Although techniques for generating target-specific molecular libraries based on RNN models are mature, research related to drug design and screening continues around the clock. Research based on de novo drug design methods to generate larger quantities of valid compounds is necessary. METHODS: In this study, a molecular generation model based on RNN was designed, which abandoned the traditional way of stacked RNN and introduced the Nested long short-term memory network structure. To enrich the library of focused molecules for specific targets, we fine-tuned the model using active molecules from novel coronavirus pneumonia and screened the molecules using machine learning models. Following rigorous screening, the selected molecules underwent molecular docking with the SARS-CoV-2 M-pro receptor using AutoDock2.4 to identify the top 3 potential inhibitors. Subsequently, 100-ns molecular dynamics simulations were conducted using Amber22. Molecule parameterization involved the GAFF2 force field, while the proteins were modeled using the ff19SB force field, with solvation facilitated by a truncated octahedral TIP3P solvent environment. Upon completion of molecular dynamics simulations, stability of ligand-protein complexes was assessed by analysis of RMSD, H-bonds, and MM-GBSA. Reasonable results prove that the model can complete the task of de novo drug design and has the potential to be ideal drug molecules.

Generating six pairs of bandwidth-expanded entangled sideband modes via time delay compensation.

Quantum entanglement is an important pillar of quantum information processing. In addition to the entanglement degree, the bandwidth of entangled states becomes another focus of quantum communication. Here, by virtue of a broadband frequency-dependent beam splitter, we experimentally demonstrate six pairs of independent entangled sideband modes with maximum entanglement degree of 8.1 dB. Utilizing a time delay compensation scheme, the bandwidth of independent entangled sideband modes is expanded to dozens of megahertz. This work provides a valuable resource to implement efficient quantum information processing.

De novo drug design based on Stack-RNN with multi-objective reward-weighted sum and reinforcement learning.

CONTEXT: In recent decades, drug development has become extremely important as different new diseases have emerged. However, drug discovery is a long and complex process with a very low success rate, and methods are needed to improve the efficiency of the process and reduce the possibility of failure. Among them, drug design from scratch has become a promising approach. Molecules are generated from scratch, reducing the reliance on trial and error and prefabricated molecular repositories, but the optimization of its molecular properties is still a challenging multi-objective optimization problem. METHODS: In this study, two stack-augmented recurrent neural networks were used to compose a generative model for generating drug-like molecules, and then reinforcement learning was used for optimization to generate molecules with desirable properties, such as binding affinity and the logarithm of the partition coefficient between octanol and water. In addition, a memory storage network was added to increase the internal diversity of the generated molecules. For multi-objective optimization, we proposed a new approach which utilized the magnitude of different attribute reward values to assign different weights to molecular optimization. The proposed model not only solves the problem that the properties of the generated molecules are extremely biased towards a certain attribute due to the possible conflict between the attributes, but also improves various properties of the generated molecules compared with the traditional weighted sum and alternating weighted sum, among which the molecular validity reaches 97.3%, the internal diversity is 0.8613, and the desirable molecules increases from 55.9 to 92%.

DeepGpgs: a novel deep learning framework for predicting arginine methylation sites combined with Gaussian prior and gated self-attention mechanism.

Protein arginine methylation is an important posttranslational modification (PTM) associated with protein functional diversity and pathological conditions including cancer. Identification of methylation binding sites facilitates a better understanding of the molecular function of proteins. Recent developments in the field of deep neural networks have led to a proliferation of deep learning-based methylation identification studies because of their fast and accurate prediction. In this paper, we propose DeepGpgs, an advanced deep learning model incorporating Gaussian prior and gated attention mechanism. We introduce a residual network channel to extract the evolutionary information of proteins. Then we combine the adaptive embedding with bidirectional long short-term memory networks to form a context-shared encoder layer. A gated multi-head attention mechanism is followed to obtain the global information about the sequence. A Gaussian prior is injected into the sequence to assist in predicting PTMs. We also propose a weighted joint loss function to alleviate the false negative problem. We empirically show that DeepGpgs improves Matthews correlation coefficient by 6.3% on the arginine methylation independent test set compared with the existing state-of-the-art methylation site prediction methods. Furthermore, DeepGpgs has good robustness in phosphorylation site prediction of SARS-CoV-2, which indicates that DeepGpgs has good transferability and the potential to be extended to other modification sites prediction. The open-source code and data of the DeepGpgs can be obtained from https://github.com/saizhou1/DeepGpgs.

PKSPS: a novel method for predicting kinase of specific phosphorylation sites based on maximum weighted bipartite matching algorithm and phosphorylation sequence enrichment analysis.

With the development of biotechnology, a large number of phosphorylation sites have been experimentally confirmed and collected, but only a few of them have kinase annotations. Since experimental methods to detect kinases at specific phosphorylation sites are expensive and accidental, some computational methods have been proposed to predict the kinase of these sites, but most methods only consider single sequence information or single functional network information. In this study, a new method Predicting Kinase of Specific Phosphorylation Sites (PKSPS) is developed to predict kinases of specific phosphorylation sites in human proteins by combining PKSPS-Net with PKSPS-Seq, which considers protein-protein interaction (PPI) network information and sequence information. For PKSPS-Net, kinase-kinase and substrate-substrate similarity are quantified based on the topological similarity of proteins in the PPI network, and maximum weighted bipartite matching algorithm is proposed to predict kinase-substrate relationship. In PKSPS-Seq, phosphorylation sequence enrichment analysis is used to analyze the similarity of local sequences around phosphorylation sites and predict the kinase of specific phosphorylation sites (KSP). PKSPS has been proved to be more effective than the PKSPS-Net or PKSPS-Seq on different sets of kinases. Further comparison results show that the PKSPS method performs better than existing methods. Finally, the case study demonstrates the effectiveness of the PKSPS in predicting kinases of specific phosphorylation sites. The open source code and data of the PKSPS can be obtained from https://github.com/guoxinyunncu/PKSPS.

Overcoming the inadaptability of sparse group lasso for data with various group structures by stacking.

MOTIVATION: Efficiently identifying genes based on gene expression level have been studied to help to classify different cancer types and improve the prediction performance. Logistic regression model based on regularization technique is often one of the effective approaches for simultaneously realizing prediction and feature (gene) selection in genomic data of high dimensionality. However, standard methods ignore biological group structure and generally result in poorer predictive models. RESULTS: In this article, we develop a classifier named Stacked SGL that satisfies the criteria of prediction, stability and selection based on sparse group lasso penalty by stacking. Sparse group lasso has a mixing parameter representing the ratio of lasso to group lasso, thus providing a compromise between selecting a subset of sparse feature groups and introducing sparsity within each group. We propose to use stacked generalization to combine different ratios rather than choosing one ratio, which could help to overcome the inadaptability of sparse group lasso for some data. Considering that stacking weakens feature selection, we perform a post hoc feature selection which might slightly reduce predictive performance, but it shows superior in feature selection. Experimental results on simulation demonstrate that our approach enjoys competitive and stable classification performance and lower false discovery rate in feature selection for varying sets of data compared with other regularization methods. In addition, our method presents better accuracy in three public cancer datasets and identifies more powerful discriminatory and potential mutation genes for thyroid carcinoma. AVAILABILITY AND IMPLEMENTATION: The real data underlying this article are available from https://github.com/huanheaha/Stacked_SGL; https://zenodo.org/record/5761577#.YbAUyciEwk2. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Entangled sideband control scheme via frequency-comb-type seed beam.

We report a control scheme of entangled sideband modes without coherent amplitude by employing a frequency-comb-type seed beam. In this scheme, each tooth of the frequency comb serves as a control field for the corresponding downconversion mode. Consequently, all the degrees of freedom can be actively controlled, and the entanglement degrees are higher than 6.7 dB for two pairs of sidebands. We believe that this scheme provides a simple solution for the control of sideband modes, which could be further applied to achieve compact channel multiplexing quantum communications.

DeepCLA: A Hybrid Deep Learning Approach for the Identification of Clathrin.

Clathrin is a highly evolutionarily conserved protein, which can affect membrane cleavage and membrane release of vesicles. The absence of clathrin in the cellular system affects a variety of human diseases. Effective recognition of clathrin plays an important role in the development of drugs to treat related diseases. In recent years, deep learning has been widely applied in the field of bioinformatics because of its high efficiency and accuracy. In this study, we propose a deep learning framework, DeepCLA, which combines two different network structures, including a convolutional neural network and a bidirectional long short-term memory network to identify clathrin. The investigation of different deep network architectures demonstrates that the prediction performance of a hybrid depth network model is better than that of a single depth network. On the independent test dataset, DeepCLA outperforms the state-of-the-art methods. It suggests that DeepCLA is an effective approach for clathrin prediction and can provide more instructive guidance for further experimental investigation of clathrin. Moreover, the source code and training data of DeepCLA are provided at https://github.com/ZhangZhang89/DeepCLA.

Demonstration of Channel Multiplexing Quantum Communication Exploiting Entangled Sideband Modes.

Channel multiplexing quantum communication based on exploiting continuous-variable entanglement of optical modes offers great potential to enhance channel capacity and save quantum resource. Here, we present a frequency-comb-type control scheme for simultaneously extracting a lot of entangled sideband modes with arbitrary frequency detuning from a squeezed state of light. We experimentally demonstrate fourfold channel multiplexing quantum dense coding communication by exploiting the extracted four pairs of entangled sideband modes. Due to high entanglement and wide frequency separation between each entangled pairs, these quantum channels have large channel capacity and the cross talking effect can be avoided. The achieved channel capacities have surpassed that of all classical and quantum communication under the same bandwidth published so far. The presented scheme can be extended to more channels if more entangled sideband modes are extracted.

Observation of a comb of squeezed states with a strong squeezing factor by a bichromatic local oscillator.

We demonstrate the experimental detection of an optical squeezing covering several higher resonances of the optical parametric amplifier (OPA) by adopting a bichromatic local oscillator (BLO). The BLO is generated from a waveguide electro-optic phase modulator (WGM) and subsequent optical mode cleaner (OMC), without the need of additional power balance and phase control. The WGM is used for generating the frequency-shifted sideband beams with equal power and certain phase difference, and the OMC is used for filtering the unwanted optical modes. Among a measurement frequency range from 0 to 16.64 GHz, the maximum squeezing factors are superior to 10 dB below the shot noise limit for the first three discrete odd-order resonances of the OPA.

Computational prediction and analysis of species-specific fungi phosphorylation via feature optimization strategy.

Protein phosphorylation is a reversible and ubiquitous post-translational modification that primarily occurs at serine, threonine and tyrosine residues and regulates a variety of biological processes. In this paper, we first briefly summarized the current progresses in computational prediction of eukaryotic protein phosphorylation sites, which mainly focused on animals and plants, especially on human, with a less extent on fungi. Since the number of identified fungi phosphorylation sites has greatly increased in a wide variety of organisms and their roles in pathological physiology still remain largely unknown, more attention has been paid on the identification of fungi-specific phosphorylation. Here, experimental fungi phosphorylation sites data were collected and most of the sites were classified into different types to be encoded with various features and trained via a two-step feature optimization method. A novel method for prediction of species-specific fungi phosphorylation-PreSSFP was developed, which can identify fungi phosphorylation in seven species for specific serine, threonine and tyrosine residues (http://computbiol.ncu.edu.cn/PreSSFP). Meanwhile, we critically evaluated the performance of PreSSFP and compared it with other existing tools. The satisfying results showed that PreSSFP is a robust predictor. Feature analyses exhibited that there have some significant differences among seven species. The species-specific prediction via two-step feature optimization method to mine important features for training could considerably improve the prediction performance. We anticipate that our study provides a new lead for future computational analysis of fungi phosphorylation.

Dependence of the squeezing and anti-squeezing factors of bright squeezed light on the seed beam power and pump beam noise.

We demonstrate the dependence of the squeezing and anti-squeezing factors on the seed beam power at different pump beam noise levels. The results indicate that a seed field injected into the optical parametric amplifier (OPA) dramatically degenerates the squeezing factor due to noise coupling between the pump and seed fields, even if both the pump and seed fields reach the shot noise limit. The squeezing and anti-squeezing factors are immune to the pump beam noise due to no noise coupling when the system operates for the generation of squeezed vacuum states. The squeezing factor degrades gradually as the pump beam intensity noise and seed beam power is increased. The influence of the two orthogonal quadrature variations is mutually independent of each other.

Proteomic analysis and prediction of amino acid variations that influence protein posttranslational modifications.

Accumulative studies have indicated that amino acid variations through changing the type of residues of the target sites or key flanking residues could directly or indirectly influence protein posttranslational modifications (PTMs) and bring about a detrimental effect on protein function. Computational mutation analysis can greatly narrow down the efforts on experimental work. To increase the utilization of current computational resources, we first provide an overview of computational prediction of amino acid variations that influence protein PTMs and their functional analysis. We also discuss the challenges that are faced while developing novel in silico approaches in the future. The development of better methods for mutation analysis-related protein PTMs will help to facilitate the development of personalized precision medicine.

Prediction and functional analysis of prokaryote lysine acetylation site by incorporating six types of features into Chou's general PseAAC.

Lysine acetylation is one of the most important types of protein post-translational modifications (PTM) that are widely involved in cellular regulatory processes. To fully understand the regulatory mechanism of acetylation, identification of acetylation sites is first and most important. However, experimental identification of protein acetylation sites is often time consuming and expensive. Thus, it is popular that predicts PTM sites by computational methods in recent years. Here, we developed a novel method, ProAcePred 2.0, to predict species-specific prokaryote lysine acetylation sites. In this study, we employed an efficient position-specific analysis strategy information gain method to constitute position-specific window of acetylation peptide, and then incorporated different types of features and adopted elastic net algorithm to optimize feature vectors for model learning. The prediction model achieved area under the receiver operating characteristic curve value of six species in training datasets, which are 0.78, 0.752, 0.783, 0.718, 0.839 and 0.826, of Escherichia coli, Corynebacterium glutamicum, Mycobacterium tuberculosis, Bacillus subtilis, S. typhimurium and Geobacillus kaustophilus, respectively. And our method was highly competitive for the majority of species when compared with other methods by using independent test datasets. In addition, function analyses demonstrated that different organisms were preferentially involved in different biological processes and pathways. The detailed analyses in this paper could help us to understand more of the acetylation mechanism and provide guidance for the related experimental validation. A user-friendly online web service of ProAcePred 2.0 can be freely available at http://computbiol.ncu.edu.cn/PAPred.

PredGly: predicting lysine glycation sites for Homo sapiens based on XGboost feature optimization.

MOTIVATION: Protein glycation is a familiar post-translational modification (PTM) which is a two-step non-enzymatic reaction. Glycation not only impairs the function but also changes the characteristics of the proteins so that it is related to many human diseases. It is still much more difficult to systematically detect glycation sites due to the glycated residues without crucial patterns. Computational approaches, which can filter supposed sites prior to experimental verification, can extremely increase the efficiency of experiment work. However, the previous lysine glycation prediction method uses a small number of training datasets. Hence, the model is not generalized or pervasive. RESULTS: By searching from a new database, we collected a large dataset in Homo sapiens. PredGly, a novel software, can predict lysine glycation sites for H.sapiens, which was developed by combining multiple features. In addition, XGboost was adopted to optimize feature vectors and to improve the model performance. Through comparing various classifiers, support vector machine achieved an optimal performance. On the basis of a new independent test set, PredGly outperformed other glycation tools. It suggests that PredGly can provide more instructive guidance for further experimental research of lysine glycation. AVAILABILITY AND IMPLEMENTATION: https://github.com/yujialinncu/PredGly. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Detection and perfect fitting of 13.2 dB squeezed vacuum states by considering green-light-induced infrared absorption.

We report on a high-level squeezed vacuum state with maximum quantum noise reduction of 13.2 dB directly detected at the pump power of 180 mW. The pump power dependence of the squeezing factor is experimentally exhibited. When considering only loss and phase fluctuation, the fitting results have a large deviation from the measurement value near the threshold. By integrating green-light-induced infrared absorption (GLIIRA) loss, the squeezing factor can be perfectly fitted in the whole pump power range. The result indicates that GLIIRA loss should be thoroughly considered and quantified in the generation of high-level squeezed states.

ProAcePred: prokaryote lysine acetylation sites prediction based on elastic net feature optimization.

Motivation: Lysine acetylation exists extensively in prokaryotes, and plays a vital role in function adjustment. Recent progresses in the identification of prokaryote acetylation substrates and sites provide a great opportunity to explore the difference of substrate site specificity between prokaryotic and eukaryotic acetylation. Motif analysis suggests that prokaryotic and eukaryotic acetylation sites have distinct location-specific difference, and it is necessary to develop a prokaryote-specific acetylation sites prediction tool. Results: Therefore, we collected nine species of prokaryote lysine acetylation data from various databases and literature, and developed a novel online tool named ProAcePred for predicting prokaryote lysine acetylation sites. Optimization of feature vectors via elastic net could considerably improve the prediction performance. Feature analyses demonstrated that evolutionary information played significant roles in prediction model for prokaryote acetylation. Comparison between our method and other tools suggested that our species-specific prediction outperformed other existing works. We expect that the ProAcePred could provide more instructive help for further experimental investigation of prokaryotes acetylation. Availability and implementation: http://computbiol.ncu.edu.cn/ProAcePred. Supplementary information: Supplementary data are available at Bioinformatics online.

Computational Prediction and Analysis for Tyrosine Post-Translational Modifications via Elastic Net.

The tyrosine residue has been identified as suffering three major post-translational modifications (PTMs) including nitration, sulfation, and phosphorylation, which could be involved in different physiological and pathological processes. Multiple tyrosine residues of the whole protein may be modified concurrently, where PTM of a single tyrosine may affect modification of other neighboring tyrosine residues. Hence, it is significant and beneficial to predict nitration, sulfation, and phosphorylation of tyrosine residues in the whole protein sequence. Here, we introduce elastic net to perform feature selection and develop a predictor named TyrPred for predicting nitrotyrosine, sulfotyrosine, and kinase-specific tyrosine phosphorylation sites on the basis of support vector machine. We critically evaluate the performance of TyrPred and compare it with other existing tools. The satisfying results show that using elastic net to mine important features for training can considerably improve the prediction performance. Feature optimization indicates that evolutionary information is significant and contributes to the prediction model. The online tool is established at http://computbiol.ncu.edu.cn/TyrPred . We anticipate that TyrPred can provide useful complements to the existing approaches in this field.