Unknown Protein Sequences Research Articles

Deep Learning-Based Self-Adaptive Evolution of Enzymes

AbstractBiocatalysis has been widely used to prepare drug leads and intermediates. Enzymatic synthesis has advantages, mainly in terms of strict chirality and regional selectivity compared with chemical methods. However, the enzymatic properties of wild-type enzymes may or may not meet the requirements for biopharmaceutical applications. Therefore, protein engineering is required to improve their catalytic activities. Thanks to advances in algorithmic models and the accumulation of immense biological data, artificial intelligence can provide novel approaches for the functional evolution of enzymes. Deep learning has the advantage of learning functions that can predict the properties of previously unknown protein sequences. Deep learning-based computational algorithms can intelligently navigate the sequence space and reduce the screening burden during evolution. Thus, intelligent computational design combined with laboratory evolution is a powerful and potentially versatile strategy for developing enzymes with novel functions. Herein, we introduce and summarize deep-learning-assisted enzyme functional adaptive evolution strategies based on recent studies on the application of deep learning in enzyme design and evolution. Altogether, with the developments of technology and the accumulation of data for the characterization of enzyme functions, artificial intelligence may become a powerful tool for the design and evolution of intelligent enzymes in the future.

Language models can identify enzymatic binding sites in protein sequences

Recent advances in language modeling have had a tremendous impact on how we handle sequential data in science. Language architectures have emerged as a hotbed of innovation and creativity in natural language processing over the last decade, and have since gained prominence in modeling proteins and chemical processes, elucidating structural relationships from textual/sequential data. Surprisingly, some of these relationships refer to three-dimensional structural features, raising important questions on the dimensionality of the information encoded within sequential data. Here, we demonstrate that the unsupervised use of a language model architecture to a language representation of bio-catalyzed chemical reactions can capture the signal at the base of the substrate-binding site atomic interactions. This allows us to identify the three-dimensional binding site position in unknown protein sequences. The language representation comprises a reaction-simplified molecular-input line-entry system (SMILES) for substrate and products, and amino acid sequence information for the enzyme. This approach can recover, with no supervision, 52.13% of the binding site when considering co-crystallized substrate-enzyme structures as ground truth, vastly outperforming other attention-based models.

Exploiting protein language models for the precise classification of ion channels and ion transporters.

This study introduces TooT-PLM-ionCT, a comprehensive framework that consolidates three distinct systems, each meticulously tailored for one of the following tasks: distinguishing ion channels (ICs) from membrane proteins (MPs), segregating ion transporters (ITs) from MPs, and differentiating ICs from ITs. Drawing upon the strengths of six Protein Language Models (PLMs)-ProtBERT, ProtBERT-BFD, ESM-1b, ESM-2 (650M parameters), and ESM-2 (15B parameters), TooT-PLM-ionCT employs a combination of traditional classifiers and deep learning models for nuanced protein classification. Originally validated on an existing dataset by previous researchers, our systems demonstrated superior performance in identifying ITs from MPs and distinguishing ICs from ITs, with the IC-MP discrimination achieving state-of-the-art results. In light of recommendations for additional validation, we introduced a new dataset, significantly enhancing the robustness and generalization of our models across bioinformatics challenges. This new evaluation underscored the effectiveness of TooT-PLM-ionCT in adapting to novel data while maintaining high classification accuracy. Furthermore, this study explores critical factors affecting classification accuracy, such as dataset balancing, the impact of using frozen versus fine-tuned PLM representations, and the variance between half and full precision in floating-point computations. To facilitate broader application and accessibility, a web server (https://tootsuite.encs.concordia.ca/service/TooT-PLM-ionCT) has been developed, allowing users to evaluate unknown protein sequences through our specialized systems for IC-MP, IT-MP, and IC-IT classification tasks.

Multilevel characterization of unknown protein sequences using hierarchical long short term memory model

Multilevel characterization of unknown protein sequences using hierarchical long short term memory model

MLapRVFL: Protein sequence prediction based on Multi-Laplacian Regularized Random Vector Functional Link

Protein sequence classification is a crucial research field in bioinformatics, playing a vital role in facilitating functional annotation, structure prediction, and gaining a deeper understanding of protein function and interactions. With the rapid development of high-throughput sequencing technologies, a vast amount of unknown protein sequence data is being generated and accumulated, leading to an increasing demand for protein classification and annotation. Existing machine learning methods still have limitations in protein sequence classification, such as low accuracy and precision of classification models, rendering them less valuable in practical applications. Additionally, these models often lack strong generalization capabilities and cannot be widely applied to various types of proteins. Therefore, accurately classifying and predicting proteins remains a challenging task. In this study, we propose a protein sequence classifier called Multi-Laplacian Regularized Random Vector Functional Link (MLapRVFL). By incorporating Multi-Laplacian and L2,1−norm regularization terms into the basic Random Vector Functional Link (RVFL) method, we effectively improve the model's generalization performance, enhance the robustness and accuracy of the classification model. The experimental results on two commonly used datasets demonstrate that MLapRVFL outperforms popular machine learning methods and achieves superior predictive performance compared to previous studies. In conclusion, the proposed MLapRVFL method makes significant contributions to protein sequence prediction.

Comparative Functional Genomics Studies for Understanding the Hypothetical Proteins in Mycobacterium Tuberculosis Variant Microti 12

Comparative Functional Genomics Studies for Understanding the Hypothetical Proteins in Mycobacterium Tuberculosis Variant Microti 12

DeepAProt: Deep learning based abiotic stress protein sequence classification and identification tool in cereals.

The impact of climate change has been alarming for the crop growth. The extreme weather conditions can stress the crops and reduce the yield of major crops belonging to Poaceae family too, that sustains 50% of the world's food calorie and 20% of protein intake. Computational approaches, such as artificial intelligence-based techniques have become the forefront of prediction-based data interpretation and plant stress responses. In this study, we proposed a novel activation function, namely, Gaussian Error Linear Unit with Sigmoid (SIELU) which was implemented in the development of a Deep Learning (DL) model along with other hyper parameters for classification of unknown abiotic stress protein sequences from crops of Poaceae family. To develop this models, data pertaining to four different abiotic stress (namely, cold, drought, heat and salinity) responsive proteins of the crops belonging to poaceae family were retrieved from public domain. It was observed that efficiency of the DL models with our proposed novel SIELU activation function outperformed the models as compared to GeLU activation function, SVM and RF with 95.11%, 80.78%, 94.97%, and 81.69% accuracy for cold, drought, heat and salinity, respectively. Also, a web-based tool, named DeepAProt (http://login1.cabgrid.res.in:5500/) was developed using flask API, along with its mobile app. This server/App will provide researchers a convenient tool, which is rapid and economical in identification of proteins for abiotic stress management in crops Poaceae family, in endeavour of higher production for food security and combating hunger, ensuring UN SDG goal 2.0.

Reliability of the In Silico Prediction Approach to In Vitro Evaluation of Bacterial Toxicity.

Several pathogens that spread through the air are highly contagious, and related infectious diseases are more easily transmitted through airborne transmission under indoor conditions, as observed during the COVID-19 pandemic. Indoor air contaminated by microorganisms, including viruses, bacteria, and fungi, or by derived pathogenic substances, can endanger human health. Thus, identifying and analyzing the potential pathogens residing in the air are crucial to preventing disease and maintaining indoor air quality. Here, we applied deep learning technology to analyze and predict the toxicity of bacteria in indoor air. We trained the ProtBert model on toxic bacterial and virulence factor proteins and applied them to predict the potential toxicity of some bacterial species by analyzing their protein sequences. The results reflect the results of the in vitro analysis of their toxicity in human cells. The in silico-based simulation and the obtained results demonstrated that it is plausible to find possible toxic sequences in unknown protein sequences.

Functional characterization of unknown protein sequences using Neuro-Fuzzy based machine learning approach and sequence augmented feature

Functional characterization of the Unknown Protein Sequence (UPS) is significant for biological research such as disease diagnosis and drug design. In this work Neuro-Fuzzy based machine learning framework is proposed with two levels of predictions, for functional characterization of UPS using augmented features and subcellular localization of the protein sequences. In the first level, Neuro-Fuzzy Approach (NFA) is applied for the categorization of UPS as bacterial or non-bacterial protein sequence. While NFA is capable to overcome the likelihood prediction problem of supervised learning algorithms with untrained samples such as UPS. In the next level functions of bacterial protein sequences are characterized using Protein Subcellular Localization (PSL) model. Physicochemical and evolutionary informations of the protein sequences are extracted and augmented as a protein feature vector that preserves the protein-residue-property and sequence-order-information. Various individual and ensemble classifiers such as Decision-Tree (C-4.5), k-Nearest-Neighbor (k-NN), Multi-Layer-Perceptron (MLP), Naïve-Bayes (NB), AdaBoost, and Gradient-Boosting-Machine (GBM) are used for the formation of the PSL model. PSL model is trained with augmented features of known Gram-Negative Bacterial Protein Sequence (GN_BPS) dataset with 10-fold cross-validation and 97.94% accuracy is achieved through C-4.5 classifier. Validated PSL model is further utilized for the functional characterization of the Unknown G- Bacterial Protein Sequences (Unk_GN_BPS) such as Unk_GN_156 and Unk_GN_61 datasets. The accuracy achieved for Unk_GN_156 is 78.20% with C-4.5 and 79.32% for the Unk_GN_61 through k-NN.

LiGIoNs: A computational method for the detection and classification of ligand-gated ion channels

Ligand-Gated Ion Channels (LGICs) is one of the largest groups of transmembrane proteins. Due to their major role in synaptic transmission, both in the nervous system and the somatic neuromuscular junction, LGICs present attractive therapeutic targets. During the last few years, several computational methods for the detection of LGICs have been developed. These methods are based on machine learning approaches utilizing features extracted solely from the amino acid composition. Here we report the development of LiGIoNs, a profile Hidden Markov Model (pHMM) method for the prediction and ligand-based classification of LGICs. The method consists of a library of 10 pHMMs, one per LGIC subfamily, built from the alignment of representative LGIC sequences. In addition, 14 Pfam pHMMs are used to further annotate and classify unknown protein sequences into one of the 10 LGIC subfamilies. Evaluation of the method showed that it outperforms existing methods in the detection of LGICs. On top of that, LiGIoNs is the only currently available method that classifies LGICs into subfamilies.The method is available online at http://bioinformatics.biol.uoa.gr/ligions/.

In Silico Study of Secondary Structure of Hemoglobin Protein

Protein structure prediction is one of the important goals in the area of bioinformatics and biotechnology. Prediction methods include structure prediction of both secondary and tertiary structures of protein. Protein secondary structure prediction infers knowledge related to presence of helixes, sheets and coils in a polypeptide chain whereas protein tertiary structure prediction infers knowledge related to three dimensional structures of proteins. Protein secondary structures represent the possible motifs or regular expressions represented as patterns that are predicted from primary protein sequence in the form of alpha helix, betastr and and coils. The secondary structure prediction is useful as it infers information related to the structure and function of unknown protein sequence. There are various secondary structure prediction methods used to predict about helixes, sheets and coils. Based on these methods there are various prediction tools under study. This study includes prediction of hemoglobin using various tools. The results produced inferred knowledge with reference to percentage of amino acids participating to produce helices, sheets and coils. PHD and DSC produced the best of the results out of all the tools used.

A Useful Tool for the Identification of DNA-Binding Proteins Using Graph Convolutional Network

Background: DNA and protein are important components of living organisms. DNA binding protein is a helicase, which is a protein specifically responsible for binding to DNA single- stranded regions. It is a necessary component for DNA replication, recombination and repair, and plays a key role in the function of various biomolecules. Although there are already some classification prediction methods for this protein, the use of graph neural networks for this work is still limited. Objective: The classification of unknown protein sequences into the correct categories, subcategories and families is important for biological sciences. In this article, using graph neural networks, we developed a novel predictor GCN-DBP for protein classification prediction. Methods: Each protein sequence is treated as a document in this study, and then segment the words according to the concept of k-mer, thereby, finally achieving the purpose of segmenting the document. This research aims to use document word relationships and word co-occurrence as a corpus to construct a text graph, and then learn protein sequence information by two-layer graph convolutional networks. Results: Finally, we tested GCN-DBP on the independent data set PDB2272, and its accuracy reached 64.17% and MCC was 28.32%. Moreover, in order to compare the proposed method with other existing methods, we have conducted more experiments. Conclusion: The results show that the proposed method is superior to the other four methods and will be a useful tool.

Long short term memory based functional characterization model for unknown protein sequences using ensemble of shallow and deep features

Long short term memory based functional characterization model for unknown protein sequences using ensemble of shallow and deep features

Augmented sequence features and subcellular localization for functional characterization of unknown protein sequences.

Advances in high-throughput techniques lead to evolving a large number of unknown protein sequences (UPS). Functional characterization of UPS is significant for the investigation of disease symptoms and drug repositioning. Protein subcellular localization is imperative for the functional characterization of protein sequences. Diverse techniques are used on protein sequences for feature extraction. However, many times a single feature extraction technique leads to poor prediction performance. In this paper, two feature augmentations are described through sequence induced, physicochemical, and evolutionary information of the amino acid residues. While augmented features preserve the sequence-order-information and protein-residue-properties. Two bacterial protein datasets Gram-Positive (G +) and Gram-Negative (G-) are utilized for the experimental work. After performing essential preprocessing on protein datasets, two sets of feature vectors are obtained. These feature vectors are used separately to train the different individual and ensembles such as decision tree (C4.5), k-nearest neighbor (k-NN), multi-layer perceptron (MLP), Naïve Bayes (NB), support vector machine (SVM), AdaBoost, gradient boosting machine (GBM), and random forest (RF) with fivefold cross-validation. Prediction results of the model demonstrate that overall accuracy reported by C4.5 is highest 99.57% on G + and 97.47% on G- datasets with known protein sequences. Similarly, for theUPS overall accuracy of G + is 85.17% with SVM and 82.45% with G- dataset using MLP.

An Approach to Enhance the Design of Protein Sequence Classifier Using Data Mining

Analysis of protein plays a major role in bio-informatics domain. Protein sequence classification is most vital in this analysis of protein. In this scenario, different researchers proposed various techniques to classify protein sequence which is reviewed here at first. Different features extraction procedure from the protein sequence is a common part of all proposed classification model. But selection and arrangement of features are the most challenging task for classifying unknown protein sequences into its known families with high accuracy and preferable computational time. Analysis of recent trend points that traditional database system fails to classify a large amount of biological data which is overcome by data mining approach. New classification model involving four phases has been proposed here for classifying the unknown sequences into its family, generating knowledge-based as well as data analysis procedure. Here a hybrid algorithm with the combination of Fuzzy ARTMAP and neural network based algorithm is used. Neighbourhood analysis is also done in the final stage. This proposed model has implemented and validated with 497 different test sequences. Finally, it is also compared with the previous model and produce high accuracy level with low computation time than previous.

2018 YPIC Challenge: A Case Study in Characterizing an Unknown Protein Sample.

For the 2018 YPIC Challenge, contestants were invited to try to decipher two unknown English questions encoded by a synthetic protein expressed in Escherichia coli. In addition to deciphering the sentence, contestants were asked to determine the three-dimensional structure and detect any post-translation modifications left by the host organism. We present our experimental and computational strategy to characterize this sample by identifying the unknown protein sequence and detecting the presence of post-translational modifications. The sample was acquired with dynamic exclusion disabled to increase the signal-to-noise ratio of the measured molecules, after which spectral clustering was used to generate high-quality consensus spectra. De novo spectrum identification was used to determine the synthetic protein sequence, and any post-translational modifications introduced by E. coli on the synthetic protein were analyzed via spectral networking. This workflow resulted in a de novo sequence coverage of 70%, on par with sequence database searching performance. Additionally, the spectral networking analysis indicated that no systematic modifications were introduced on the synthetic protein by E. coli. The strategy presented here can be directly used to analyze samples for which no protein sequence information is available or when the identity of the sample is unknown. All software and code to perform the bioinformatics analysis is available as open source, and self-contained Jupyter notebooks are provided to fully recreate the analysis.

VaxiJen Dataset of Bacterial Immunogens: An Update.

Identifying immunogenic proteins is the first stage in vaccine design and development. VaxiJen is the most widely used and highly cited server for immunogenicity prediction. As the developers of VaxiJen, we are obliged to update and improve it regularly. Here, we present an updated dataset of bacterial immunogens containing 317 experimentally proven immunogenic proteins of bacterial origin, of which 60% have been reported during the last 10 years. PubMed was searched for papers containing data for novel immunogenic proteins tested on humans till March 2017. Corresponding protein sequences were collected from NCBI and UniProtKB. The set was curated manually for multiple protein fragments, isoforms, and duplicates. The final curated dataset consists of 306 immunogenic proteins tested on humans derived from 47 bacterial microorganisms. Certain proteins have several isoforms. All were considered, and the total protein sequences in the set are 317. The updated set contains 206 new immunogens, compared to the previous VaxiJen bacterial dataset. The average number of immunogens per species is 6.7. The set also contains 12 fusion proteins and 41 peptide fragments and epitopes. The dataset includes the names of bacterial microorganisms, protein names, and protein sequences in FASTA format. Currently, the updated VaxiJen bacterial dataset is the best known manually-curated compilation of bacterial immunogens. It is freely available at http://www.ddg-pharmfac.net/vaxi jen/dataset. It can easily be downloaded, searched, and processed. When combined with an appropriate negative dataset, this update could also serve as a training set, allowing enhanced prediction of the potential immunogenicity of unknown protein sequences.

Object-Oriented Analysis for Design of Protein Fingerprints

A new approach, based on object-oriented programming, for search and classification of protein patterns in a protein database is developed. Aiming at an improvement of the analysis of protein sequences and structures, certain basic patterns - such as motifs and fingerprints - are used. An unknown protein sequence can be classified by searching for those fingerprints from the database, assessing matches by estimating the statistical significance. The current implementation of the model algorithm for searching is a powerful tool that uses the PRINTS database as an example, however it does not support the option of adding of new features due to the conservative design of the program and the lack of publicly available code. A new version of the PRINTS database has recently been developed and this will require adding new features in the future. A novel object-oriented model for the implementation of the algorithm is proposed. This model is used to build a web application prototype, written in Python - the most widely used programming language in bioinformatics at present. The result of this study is a maintainable software with open source code that can easily be extended with new functionalities.

Mathematical Basis of Predicting Dominant Function in Protein Sequences by a Generic HMM\u2013ANN Algorithm

The accurate annotation of an unknown protein sequence depends on extant data of template sequences. This could be empirical or sets of reference sequences, and provides an exhaustive pool of probable functions. Individual methods of predicting dominant function possess shortcomings such as varying degrees of inter-sequence redundancy, arbitrary domain inclusion thresholds, heterogeneous parameterization protocols, and ill-conditioned input channels. Here, I present a rigorous theoretical derivation of various steps of a generic algorithm that integrates and utilizes several statistical methods to predict the dominant function in unknown protein sequences. The accompanying mathematical proofs, interval definitions, analysis, and numerical computations presented are meant to offer insights not only into the specificity and accuracy of predictions, but also provide details of the operatic mechanisms involved in the integration and its ensuing rigor. The algorithm uses numerically modified raw hidden markov model scores of well defined sets of training sequences and clusters them on the basis of known function. The results are then fed into an artificial neural network, the predictions of which can be refined using the available data. This pipeline is trained recursively and can be used to discern the dominant principal function, and thereby, annotate an unknown protein sequence. Whilst, the approach is complex, the specificity of the final predictions can benefit laboratory workers design their experiments with greater confidence.

NeuroPP: A Tool for the Prediction of Neuropeptide Precursors Based on Optimal Sequence Composition.

Neuropeptides (NPs) are short secreted peptides produced mainly in the nervous system and digestive system. They activate signaling cascades to control a wide range of biological functions, such as metabolism, sensation, and behavior. NPs are typically produced from a larger NP precursor (NPP) which includes a signal peptide sequence, one or more NP sequences, and other sequences. With the drastic growth of unknown protein sequences generated in the post-genomic age, it is highly desired to develop computational methods for identifying NPP rapidly and efficiently. In this article, we developed a predictor for NPPs based on optimized sequence composition of single amino acid, dipeptide, and tripeptide. Evaluated with independent data set, the predictor showed excellent performance that achieved an accuracy of 88.65% with AUC of 0.95. The corresponding web server was developed, which is freely available at http://i.uestc.edu.cn/neuropeptide/neuropp/home.html . It can help relevant researchers to screen candidate NP precursor, shorten experimental cycle, and reduce costs.