Continuous Distributed Representation of Biological Sequences for Deep Proteomics and Genomics

Discovering genomic islands in unannotated bacterial genomes using sequence embedding

Motivation

Genomic islands (GEIs) are clusters of genes in bacterial genomes that are typically acquired by horizontal gene transfer. GEIs play a crucial role in the evolution of bacteria by rapidly introducing genetic diversity and thus helping them adapt to changing environments. Specifically of interest to human health, many GEIs contain pathogenicity and antimicrobial resistance genes. Detecting GEIs is, therefore, an important problem in biomedical and environmental research. There have been many previous studies for computationally identifying GEIs. Still, most of these studies rely on detecting anomalies in the unannotated nucleotide sequences or on a fixed set of known features on annotated nucleotide sequences.

Results

Here, we present TreasureIsland, which uses a new unsupervised representation of DNA sequences to predict GEIs. We developed a high-precision boundary detection method featuring an incremental fine-tuning of GEI borders, and we evaluated the accuracy of this framework using a new comprehensive reference dataset, Benbow. We show that TreasureIsland's accuracy rivals other GEI predictors, enabling efficient and faster identification of GEIs in unannotated bacterial genomes.

Availability and implementation

TreasureIsland is available under an MIT license at: https://github.com/FriedbergLab/GenomicIslandPrediction.

SOFB is a comprehensive ensemble deep learning approach for elucidating and characterizing protein-nucleic-acid-binding residues

Proteins and nucleic-acids are essential components of living organisms that interact in critical cellular processes. Accurate prediction of nucleic acid-binding residues in proteins can contribute to a better understanding of protein function. However, the discrepancy between protein sequence information and obtained structural and functional data renders most current computational models ineffective. Therefore, it is vital to design computational models based on protein sequence information to identify nucleic acid binding sites in proteins. Here, we implement an ensemble deep learning model-based nucleic-acid-binding residues on proteins identification method, called SOFB, which characterizes protein sequences by learning the semantics of biological dynamics contexts, and then develop an ensemble deep learning-based sequence network to learn feature representation and classification by explicitly modeling dynamic semantic information. Among them, the language learning model, which is constructed from natural language to biological language, captures the underlying relationships of protein sequences, and the ensemble deep learning-based sequence network consisting of different convolutional layers together with Bi-LSTM refines various features for optimal performance. Meanwhile, to address the imbalanced issue, we adopt ensemble learning to train multiple models and then incorporate them. Our experimental results on several DNA/RNA nucleic-acid-binding residue datasets demonstrate that our proposed model outperforms other state-of-the-art methods. In addition, we conduct an interpretability analysis of the identified nucleic acid binding residue sequences based on the attention weights of the language learning model, revealing novel insights into the dynamic semantic information that supports the identified nucleic acid binding residues. SOFB is available at https://github.com/Encryptional/SOFB and https://figshare.com/articles/online_resource/SOFB_figshare_rar/25499452 .

Nextflow vs. plain bash: different approaches to the parallelization of SNP calling from the whole genome sequence data

This study compared computational approaches to parallelization of an SNP calling workflow. The data comprised DNA from five Holstein-Friesian cows sequenced with the Illumina platform. The pipeline consisted of quality control, alignment to the reference genome, post-alignment, and SNP calling. Three approaches to parallelization were compared: (i) a plain Bash script in which a pipeline for each cow was executed as separate processes invoked at the same time, (ii) a Bash script wrapped in a single Nextflow process and (iii) a Nextflow script with each component of the pipeline defined as a separate process. The results demonstrated that on average, the multi-process Nextflow script performed 15-27% faster depending on the number of assigned threads, with the biggest execution time advantage over the plain Bash approach observed with 10 threads. In terms of RAM usage, the most substantial variation was observed for the multi-process Nextflow, for which it increased with the number of assigned threads, while RAM consumption of the other setups did not depend much on the number of threads assigned for computations. Due to intermediate and log files generated, disk usage was markedly higher for the multi-process Nextflow than for the plain Bash and for the single-process Nextflow.

Prediction of Protein-Protein Interactions Based on Integrating Deep Learning and Feature Fusion

Understanding protein-protein interactions (PPIs) helps to identify protein functions and develop other important applications such as drug preparation and protein-disease relationship identification. Deep-learning-based approaches are being intensely researched for PPI determination to reduce the cost and time of previous testing methods. In this work, we integrate deep learning with feature fusion, harnessing the strengths of both approaches, handcrafted features, and protein sequence embedding. The accuracies of the proposed model using five-fold cross-validation on Yeast core and Human datasets are 96.34% and 99.30%, respectively. In the task of predicting interactions in important PPI networks, our model correctly predicted all interactions in one-core, Wnt-related, and cancer-specific networks. The experimental results on cross-species datasets, including Caenorhabditis elegans, Helicobacter pylori, Homo sapiens, Mus musculus, and Escherichia coli, also show that our feature fusion method helps increase the generalization capability of the PPI prediction model.

Antibody design using deep learning: from sequence and structure design to affinity maturation

Deep learning has achieved impressive results in various fields such as computer vision and natural language processing, making it a powerful tool in biology. Its applications now encompass cellular image classification, genomic studies and drug discovery. While drug development traditionally focused deep learning applications on small molecules, recent innovations have incorporated it in the discovery and development of biological molecules, particularly antibodies. Researchers have devised novel techniques to streamline antibody development, combining in vitro and in silico methods. In particular, computational power expedites lead candidate generation, scaling and potential antibody development against complex antigens. This survey highlights significant advancements in protein design and optimization, specifically focusing on antibodies. This includes various aspects such as design, folding, antibody-antigen interactions docking and affinity maturation.

HLAIImaster: a deep learning method with adaptive domain knowledge predicts HLA II neoepitope immunogenic responses

While significant strides have been made in predicting neoepitopes that trigger autologous CD4+ T cell responses, accurately identifying the antigen presentation by human leukocyte antigen (HLA) class II molecules remains a challenge. This identification is critical for developing vaccines and cancer immunotherapies. Current prediction methods are limited, primarily due to a lack of high-quality training epitope datasets and algorithmic constraints. To predict the exogenous HLA class II-restricted peptides across most of the human population, we utilized the mass spectrometry data to profile >223 000 eluted ligands over HLA-DR, -DQ, and -DP alleles. Here, by integrating these data with peptide processing and gene expression, we introduce HLAIImaster, an attention-based deep learning framework with adaptive domain knowledge for predicting neoepitope immunogenicity. Leveraging diverse biological characteristics and our enhanced deep learning framework, HLAIImaster is significantly improved against existing tools in terms of positive predictive value across various neoantigen studies. Robust domain knowledge learning accurately identifies neoepitope immunogenicity, bridging the gap between neoantigen biology and the clinical setting and paving the way for future neoantigen-based therapies to provide greater clinical benefit. In summary, we present a comprehensive exploitation of the immunogenic neoepitope repertoire of cancers, facilitating the effective development of "just-in-time" personalized vaccines.

Protein sequence analysis in the context of drug repurposing

MOTIVATION

Drug repurposing speeds up the development of new treatments, being less costly, risky, and time consuming than de novo drug discovery. There are numerous biological elements that contribute to the development of diseases and, as a result, to the repurposing of drugs.

METHODS

In this article, we analysed the potential role of protein sequences in drug repurposing scenarios. For this purpose, we embedded the protein sequences by performing four state of the art methods and validated their capacity to encapsulate essential biological information through visualization. Then, we compared the differences in sequence distance between protein-drug target pairs of drug repurposing and non - drug repurposing data. Thus, we were able to uncover patterns that define protein sequences in repurposing cases.

RESULTS

We found statistically significant sequence distance differences between protein pairs in the repurposing data and the rest of protein pairs in non-repurposing data. In this manner, we verified the potential of using numerical representations of sequences to generate repurposing hypotheses in the future.

Genotype sampling for deep-learning assisted experimental mapping of a combinatorially complete fitness landscape

MOTIVATION

Experimental characterization of fitness landscapes, which map genotypes onto fitness, is important for both evolutionary biology and protein engineering. It faces a fundamental obstacle in the astronomical number of genotypes whose fitness needs to be measured for any one protein. Deep learning may help to predict the fitness of many genotypes from a smaller neural network training sample of genotypes with experimentally measured fitness. Here I use a recently published experimentally mapped fitness landscape of more than 260 000 protein genotypes to ask how such sampling is best performed.

RESULTS

I show that multilayer perceptrons, recurrent neural networks, convolutional networks, and transformers, can explain more than 90% of fitness variance in the data. In addition, 90% of this performance is reached with a training sample comprising merely ≈103 sequences. Generalization to unseen test data is best when training data is sampled randomly and uniformly, or sampled to minimize the number of synonymous sequences. In contrast, sampling to maximize sequence diversity or codon usage bias reduces performance substantially. These observations hold for more than one network architecture. Simple sampling strategies may perform best when training deep learning neural networks to map fitness landscapes from experimental data.

AVAILABILITY AND IMPLEMENTATION

The fitness landscape data analyzed here is publicly available as described previously (Papkou et al. 2023). All code used to analyze this landscape is publicly available at https://github.com/andreas-wagner-uzh/fitness_landscape_sampling.

ntEmbd: Deep learning embedding for nucleotide sequences

Enabled by the explosion of data and substantial increase in computational power, deep learning has transformed fields such as computer vision and natural language processing (NLP) and it has become a successful method to be applied to many transcriptomic analysis tasks. A core advantage of deep learning is its inherent capability to incorporate feature computation within the machine learning models. This results in a comprehensive and machine-readable representation of sequences, facilitating the downstream classification and clustering tasks. Compared to machine translation problems in NLP, feature embedding is particularly challenging for transcriptomic studies as the sequences are string of thousands of nucleotides in length, which make the long-term dependencies between features from different parts of the sequence even more difficult to capture. This highlights the need for nucleotide sequence embedding methods that are capable of learning input sequence features implicitly. Here we introduce ntEmbd, a deep learning embedding tool that captures dependencies between different features of the sequences and learns a latent representation for given nucleotide sequences. We further provide two sample use cases, describing how learned RNA features can be used in downstream analysis. The first use case demonstrates ntEmbd's utility in classifying coding and noncoding RNA benchmarked against existing tools, and the second one explores the utility of learned representations in identifying adapter sequences in nanopore RNA-seq reads. The tool as well as the trained models are freely available on GitHub at https://github.com/bcgsc/ntEmbd.

Classifying alkaliphilic proteins using embeddings from protein language model

Alkaliphilic proteins have great potential as biocatalysts in biotechnology, especially for enzyme engineering. Extensive research has focused on exploring the enzymatic potential of alkaliphiles and characterizing alkaliphilic proteins. However, the current method employed for identifying these proteins that requires web lab experiment is time-consuming, labor-intensive, and expensive. Therefore, the development of a computational method for alkaliphilic protein identification would be invaluable for protein engineering and design. In this study, we present a novel approach that uses embeddings from a protein language model called ESM-2(3B) in a deep learning framework to classify alkaliphilic and non-alkaliphilic proteins. To our knowledge, this is the first attempt to employ embeddings from a pre-trained protein language model to classify alkaliphilic protein. A reliable dataset comprising 1,002 alkaliphilic and 1,866 non-alkaliphilic proteins was constructed for training and testing the proposed model. The proposed model, dubbed ALPACA, achieves performance scores of 0.88, 0.84, and 0.75 for accuracy, f1-score, and Matthew correlation coefficient respectively on independent dataset. ALPACA is likely to serve as a valuable resource for exploring protein alkalinity and its role in protein design and engineering.

Comparison of deep learning models with simple method to assess the problem of antimicrobial peptides prediction

Antibiotic-resistant strains are an emerging threat to public health. The usage of antimicrobial peptides (AMPs) is one of the promising approaches to solve this problem. For the development of new AMPs, it is necessary to have reliable prediction methods. Recently, deep learning approaches have been used to predict AMP. In this paper, we want to compare simple and complex methods for these purposes. We used the BERT transformer to create sequence embeddings and the multilayer perceptron (MLP) and light attention (LA) approaches for classification. One of them reached about 80 % accuracy and specificity in benchmark testing, which is on par with the best available methods. For comparison, we proposed a simple method using only the amino acid composition of proteins or peptides. This method has shown good results, at the level of the best methods. We have prepared a special server for predicting the ability of AMPs by amino acid composition: http://bioproteom.protres.ru/antimicrob/.

An Explainable Deep Learning Classifier of Bovine Mastitis Based on Whole-Genome Sequence Data-Circumventing the p >> n Problem

The serious drawback underlying the biological annotation of whole-genome sequence data is the p >> n problem, which means that the number of polymorphic variants (p) is much larger than the number of available phenotypic records (n). We propose a way to circumvent the problem by combining a LASSO logistic regression with deep learning to classify cows as susceptible or resistant to mastitis, based on single nucleotide polymorphism (SNP) genotypes. Among several architectures, the one with 204,642 SNPs was selected as the best. This architecture was composed of two layers with, respectively, 7 and 46 units per layer implementing respective drop-out rates of 0.210 and 0.358. The classification of the test data resulted in AUC = 0.750, accuracy = 0.650, sensitivity = 0.600, and specificity = 0.700. Significant SNPs were selected based on the SHapley Additive exPlanation (SHAP). As a final result, one GO term related to the biological process and thirteen GO terms related to molecular function were significantly enriched in the gene set that corresponded to the significant SNPs. Our findings revealed that the optimal approach can correctly predict susceptibility or resistance status for approximately 65% of cows. Genes marked by the most significant SNPs are related to the immune response and protein synthesis.

Integrated clinical and genomic models using machine-learning methods to predict the efficacy of paclitaxel-based chemotherapy in patients with advanced gastric cancer

BACKGROUND

Paclitaxel is commonly used as a second-line therapy for advanced gastric cancer (AGC). The decision to proceed with second-line chemotherapy and select an appropriate regimen is critical for vulnerable patients with AGC progressing after first-line chemotherapy. However, no predictive biomarkers exist to identify patients with AGC who would benefit from paclitaxel-based chemotherapy.

METHODS

This study included 288 patients with AGC receiving second-line paclitaxel-based chemotherapy between 2017 and 2022 as part of the K-MASTER project, a nationwide government-funded precision medicine initiative. The data included clinical (age [young-onset vs. others], sex, histology [intestinal vs. diffuse type], prior trastuzumab use, duration of first-line chemotherapy), and genomic factors (pathogenic or likely pathogenic variants). Data were randomly divided into training and validation sets (0.8:0.2). Four machine learning (ML) methods, namely random forest (RF), logistic regression (LR), artificial neural network (ANN), and ANN with genetic embedding (ANN with GE), were used to develop the prediction model and validated in the validation sets.

RESULTS

The median patient age was 64 years (range 25-91), and 65.6% of those were male. A total of 288 patients were divided into the training (n = 230) and validation (n = 58) sets. No significant differences existed in baseline characteristics between the training and validation sets. In the training set, the areas under the ROC curves (AUROC) for predicting better progression-free survival (PFS) with paclitaxel-based chemotherapy were 0.499, 0.679, 0.618, and 0.732 in the RF, LR, ANN, and ANN with GE models, respectively. The ANN with the GE model that achieved the highest AUROC recorded accuracy, sensitivity, specificity, and F1-score performance of 0.458, 0.912, 0.724, and 0.579, respectively. In the validation set, the ANN with GE model predicted that paclitaxel-sensitive patients had significantly longer PFS (median PFS 7.59 vs. 2.07 months, P = 0.020) and overall survival (OS) (median OS 14.70 vs. 7.50 months, P = 0.008). The LR model predicted that paclitaxel-sensitive patients showed a trend for longer PFS (median PFS 6.48 vs. 2.33 months, P = 0.078) and OS (median OS 12.20 vs. 8.61 months, P = 0.099).

CONCLUSIONS

These ML models, integrated with clinical and genomic factors, offer the possibility to help identify patients with AGC who may benefit from paclitaxel chemotherapy.

Protein feature engineering framework for AMPylation site prediction

AMPylation is a biologically significant yet understudied post-translational modification where an adenosine monophosphate (AMP) group is added to Tyrosine and Threonine residues primarily. While recent work has illuminated the prevalence and functional impacts of AMPylation, experimental identification of AMPylation sites remains challenging. Computational prediction techniques provide a faster alternative approach. The predictive performance of machine learning models is highly dependent on the features used to represent the raw amino acid sequences. In this work, we introduce a novel feature extraction pipeline to encode the key properties relevant to AMPylation site prediction. We utilize a recently published dataset of curated AMPylation sites to develop our feature generation framework. We demonstrate the utility of our extracted features by training various machine learning classifiers, on various numerical representations of the raw sequences extracted with the help of our framework. Tenfold cross-validation is used to evaluate the model's capability to distinguish between AMPylated and non-AMPylated sites. The top-performing set of features extracted achieved MCC score of 0.58, Accuracy of 0.8, AUC-ROC of 0.85 and F1 score of 0.73. Further, we elucidate the behaviour of the model on the set of features consisting of monogram and bigram counts for various representations using SHapley Additive exPlanations.

KEGG orthology prediction of bacterial proteins using natural language processing

BACKGROUND

The advent of high-throughput technologies has led to an exponential increase in uncharacterized bacterial protein sequences, surpassing the capacity of manual curation. A large number of bacterial protein sequences remain unannotated by Kyoto Encyclopedia of Genes and Genomes (KEGG) orthology, making it necessary to use auto annotation tools. These tools are now indispensable in the biological research landscape, bridging the gap between the vastness of unannotated sequences and meaningful biological insights.

RESULTS

In this work, we propose a novel pipeline for KEGG orthology annotation of bacterial protein sequences that uses natural language processing and deep learning. To assess the effectiveness of our pipeline, we conducted evaluations using the genomes of two randomly selected species from the KEGG database. In our evaluation, we obtain competitive results on precision, recall, and F1 score, with values of 0.948, 0.947, and 0.947, respectively.

CONCLUSIONS

Our experimental results suggest that our pipeline demonstrates performance comparable to traditional methods and excels in identifying distant relatives with low sequence identity. This demonstrates the potential of our pipeline to significantly improve the accuracy and comprehensiveness of KEGG orthology annotation, thereby advancing our understanding of functional relationships within biological systems.

Integrated Computational Pipeline for the High-Throughput Discovery of Cell Adhesion Peptides

Cell adhesion peptides (CAPs) often play a critical role in tissue engineering research. However, the discovery of novel CAPs for diverse applications remains a challenging and time-intensive process. This study presents an efficient computational pipeline integrating sequence embeddings, binding predictors, and molecular dynamics simulations to expedite the discovery of new CAPs. A Pro2vec model, trained on vast CAP data sets, was built to identify RGD-similar tripeptide candidates. These candidates were further evaluated for their binding affinity with integrin receptors using the Mutabind2 machine learning model. Additionally, molecular dynamics simulations were applied to model receptor-peptide interactions and calculate their binding free energies, providing a quantitative assessment of the binding strength for further screening. The resulting peptide demonstrated performance comparable to that of RGD in endothelial cell adhesion and spreading experimental assays, validating the efficacy of the integrated computational pipeline.

PRIEST: predicting viral mutations with immune escape capability of SARS-CoV-2 using temporal evolutionary information

The dynamic evolution of the severe acute respiratory syndrome coronavirus 2 virus is primarily driven by mutations in its genetic sequence, culminating in the emergence of variants with increased capability to evade host immune responses. Accurate prediction of such mutations is fundamental in mitigating pandemic spread and developing effective control measures. This study introduces a robust and interpretable deep-learning approach called PRIEST. This innovative model leverages time-series viral sequences to foresee potential viral mutations. Our comprehensive experimental evaluations underscore PRIEST's proficiency in accurately predicting immune-evading mutations. Our work represents a substantial step in utilizing deep-learning methodologies for anticipatory viral mutation analysis and pandemic response.

Scoring alignments by embedding vector similarity

Sequence similarity is of paramount importance in biology, as similar sequences tend to have similar function and share common ancestry. Scoring matrices, such as PAM or BLOSUM, play a crucial role in all bioinformatics algorithms for identifying similarities, but have the drawback that they are fixed, independent of context. We propose a new scoring method for amino acid similarity that remedies this weakness, being contextually dependent. It relies on recent advances in deep learning architectures that employ self-supervised learning in order to leverage the power of enormous amounts of unlabelled data to generate contextual embeddings, which are vector representations for words. These ideas have been applied to protein sequences, producing embedding vectors for protein residues. We propose the E-score between two residues as the cosine similarity between their embedding vector representations. Thorough testing on a wide variety of reference multiple sequence alignments indicate that the alignments produced using the new $E$-score method, especially ProtT5-score, are significantly better than those obtained using BLOSUM matrices. The new method proposes to change the way alignments are computed, with far-reaching implications in all areas of textual data that use sequence similarity. The program to compute alignments based on various $E$-scores is available as a web server at e-score.csd.uwo.ca. The source code is freely available for download from github.com/lucian-ilie/E-score.

Precise prediction of phase-separation key residues by machine learning

Understanding intracellular phase separation is crucial for deciphering transcriptional control, cell fate transitions, and disease mechanisms. However, the key residues, which impact phase separation the most for protein phase separation function have remained elusive. We develop PSPHunter, which can precisely predict these key residues based on machine learning scheme. In vivo and in vitro validations demonstrate that truncating just 6 key residues in GATA3 disrupts phase separation, enhancing tumor cell migration and inhibiting growth. Glycine and its motifs are enriched in spacer and key residues, as revealed by our comprehensive analysis. PSPHunter identifies nearly 80% of disease-associated phase-separating proteins, with frequent mutated pathological residues like glycine and proline often residing in these key residues. PSPHunter thus emerges as a crucial tool to uncover key residues, facilitating insights into phase separation mechanisms governing transcriptional control, cell fate transitions, and disease development.

Towards explainable interaction prediction: Embedding biological hierarchies into hyperbolic interaction space

Given the prolonged timelines and high costs associated with traditional approaches, accelerating drug development is crucial. Computational methods, particularly drug-target interaction prediction, have emerged as efficient tools, yet the explainability of machine learning models remains a challenge. Our work aims to provide more interpretable interaction prediction models using similarity-based prediction in a latent space aligned to biological hierarchies. We investigated integrating drug and protein hierarchies into a joint-embedding drug-target latent space via embedding regularization by conducting a comparative analysis between models employing traditional flat Euclidean vector spaces and those utilizing hyperbolic embeddings. Besides, we provided a latent space analysis as an example to show how we can gain visual insights into the trained model with the help of dimensionality reduction. Our results demonstrate that hierarchy regularization improves interpretability without compromising predictive performance. Furthermore, integrating hyperbolic embeddings, coupled with regularization, enhances the quality of the embedded hierarchy trees. Our approach enables a more informed and insightful application of interaction prediction models in drug discovery by constructing an interpretable hyperbolic latent space, simultaneously incorporating drug and target hierarchies and pairing them with available interaction information. Moreover, compatible with pairwise methods, the approach allows for additional transparency through existing explainable AI solutions.

Optimised stacked machine learning algorithms for genomics and genetics disorder detection in the healthcare industry

With recent advances in precision medicine and healthcare computing, there is an enormous demand for developing machine learning algorithms in genomics to enhance the rapid analysis of disease disorders. Technological advancement in genomics and imaging provides clinicians with enormous amounts of data, but prediction is still mostly subjective, resulting in problematic medical treatment. Machine learning is being employed in several domains of the healthcare sector, encompassing clinical research, early disease identification, and medicinal innovation with a historical perspective. The main objective of this study is to detect patients who, based on several medical standards, are more susceptible to having a genetic disorder. A genetic disease prediction algorithm was employed, leveraging the patient's health history to evaluate the probability of diagnosing a genetic disorder. We developed a computationally efficient machine learning approach to predict the overall lifespan of patients with a genomics disorder and to classify and predict patients with a genetic disease. The SVM, RF, and ETC are stacked using two-layer meta-estimators to develop the proposed model. The first layer comprises all the baseline models employed to predict the outcomes based on the dataset. The second layer comprises a component known as a meta-classifier. Results from the experiment indicate that the model achieved an accuracy of 90.45% and a recall score of 90.19%. The area under the curve (AUC) for mitochondrial diseases is 98.1%; for multifactorial diseases, it is 97.5%; and for single-gene inheritance, it is 98.8%. The proposed approach presents a novel method for predicting patient prognosis in a manner that is unbiased, accurate, and comprehensive. The proposed approach outperforms human professionals using the current clinical standard for genetic disease classification in terms of identification accuracy. The implementation of stacked will significantly improve the field of biomedical research by improving the anticipation of genetic diseases.

Large language models improve annotation of prokaryotic viral proteins

Viral genomes are poorly annotated in metagenomic samples, representing an obstacle to understanding viral diversity and function. Current annotation approaches rely on alignment-based sequence homology methods, which are limited by the paucity of characterized viral proteins and divergence among viral sequences. Here we show that protein language models can capture prokaryotic viral protein function, enabling new portions of viral sequence space to be assigned biologically meaningful labels. When applied to global ocean virome data, our classifier expanded the annotated fraction of viral protein families by 29%. Among previously unannotated sequences, we highlight the identification of an integrase defining a mobile element in marine picocyanobacteria and a capsid protein that anchors globally widespread viral elements. Furthermore, improved high-level functional annotation provides a means to characterize similarities in genomic organization among diverse viral sequences. Protein language models thus enhance remote homology detection of viral proteins, serving as a useful complement to existing approaches.

DeepPPThermo: A Deep Learning Framework for Predicting Protein Thermostability Combining Protein-Level and Amino Acid-Level Features

Using wet experimental methods to discover new thermophilic proteins or improve protein thermostability is time-consuming and expensive. Machine learning methods have shown powerful performance in the study of protein thermostability in recent years. However, how to make full use of multiview sequence information to predict thermostability effectively is still a challenge. In this study, we proposed a deep learning-based classifier named DeepPPThermo that fuses features of classical sequence features and deep learning representation features for classifying thermophilic and mesophilic proteins. In this model, deep neural network (DNN) and bi-long short-term memory (Bi-LSTM) are used to mine hidden features. Furthermore, local attention and global attention mechanisms give different importance to multiview features. The fused features are fed to a fully connected network classifier to distinguish thermophilic and mesophilic proteins. Our model is comprehensively compared with advanced machine learning algorithms and deep learning algorithms, proving that our model performs better. We further compare the effects of removing different features on the classification results, demonstrating the importance of each feature and the robustness of the model. Our DeepPPThermo model can be further used to explore protein diversity, identify new thermophilic proteins, and guide directed mutations of mesophilic proteins.

Decrypting orphan GPCR drug discovery via multitask learning

The drug discovery of G protein-coupled receptors (GPCRs) superfamily using computational models is often limited by the availability of protein three-dimensional (3D) structures and chemicals with experimentally measured bioactivities. Orphan GPCRs without known ligands further complicate the process. To enable drug discovery for human orphan GPCRs, multitask models were proposed for predicting half maximal effective concentrations (EC50) of the pairs of chemicals and GPCRs. Protein multiple sequence alignment features, and physicochemical properties and fingerprints of chemicals were utilized to encode the protein and chemical information, respectively. The protein features enabled the transfer of data-rich GPCRs to orphan receptors and the transferability based on the similarity of protein features. The final model was trained using both agonist and antagonist data from 200 GPCRs and showed an excellent mean squared error (MSE) of 0.24 in the validation dataset. An independent test using the orphan dataset consisting of 16 receptors associated with less than 8 bioactivities showed a reasonably good MSE of 1.51 that can be further improved to 0.53 by considering the transferability based on protein features. The informative features were identified and mapped to corresponding 3D structures to gain insights into the mechanism of GPCR-ligand interactions across the GPCR family. The proposed method provides a novel perspective on learning ligand bioactivity within the diverse human GPCR superfamily and can potentially accelerate the discovery of therapeutic agents for orphan GPCRs.

POOE: predicting oomycete effectors based on a pre-trained large protein language model

Oomycetes are fungus-like eukaryotic microorganisms which can cause catastrophic diseases in many plants. Successful infection of oomycetes depends highly on their effector proteins that are secreted into plant cells to subvert plant immunity. Thus, systematic identification of effectors from the oomycete proteomes remains an initial but crucial step in understanding plant-pathogen relationships. However, the number of experimentally identified oomycete effectors is still limited. Currently, only a few bioinformatics predictors exist to detect potential effectors, and their prediction performance needs to be improved. Here, we used the sequence embeddings from a pre-trained large protein language model (ProtTrans) as input and developed a support vector machine-based method called POOE for predicting oomycete effectors. POOE could achieve a highly accurate performance with an area under the precision-recall curve of 0.804 (area under the receiver operating characteristic curve = 0.893, accuracy = 0.874, precision = 0.777, recall = 0.684, and specificity = 0.936) in the fivefold cross-validation, considerably outperforming various combinations of popular machine learning algorithms and other commonly used sequence encoding schemes. A similar prediction performance was also observed in the independent test. Compared with the existing oomycete effector prediction methods, POOE provided very competitive and promising performance, suggesting that ProtTrans effectively captures rich protein semantic information and dramatically improves the prediction task. We anticipate that POOE can accelerate the identification of oomycete effectors and provide new hints to systematically understand the functional roles of effectors in plant-pathogen interactions. The web server of POOE is freely accessible at http://zzdlab.com/pooe/index.php. The corresponding source codes and data sets are also available at https://github.com/zzdlabzm/POOE.IMPORTANCEIn this work, we use the sequence representations from a pre-trained large protein language model (ProtTrans) as input and develop a Support Vector Machine-based method called POOE for predicting oomycete effectors. POOE could achieve a highly accurate performance in the independent test set, considerably outperforming existing oomycete effector prediction methods. We expect that this new bioinformatics tool will accelerate the identification of oomycete effectors and further guide the experimental efforts to interrogate the functional roles of effectors in plant-pathogen interaction.

metaProbiotics: a tool for mining probiotic from metagenomic binning data based on a language model

Beneficial bacteria remain largely unexplored. Lacking systematic methods, understanding probiotic community traits becomes challenging, leading to various conclusions about their probiotic effects among different publications. We developed language model-based metaProbiotics to rapidly detect probiotic bins from metagenomes, demonstrating superior performance in simulated benchmark datasets. Testing on gut metagenomes from probiotic-treated individuals, it revealed the probioticity of intervention strains-derived bins and other probiotic-associated bins beyond the training data, such as a plasmid-like bin. Analyses of these bins revealed various probiotic mechanisms and bai operon as probiotic Ruminococcaceae's potential marker. In different health-disease cohorts, these bins were more common in healthy individuals, signifying their probiotic role, but relevant health predictions based on the abundance profiles of these bins faced cross-disease challenges. To better understand the heterogeneous nature of probiotics, we used metaProbiotics to construct a comprehensive probiotic genome set from global gut metagenomic data. Module analysis of this set shows that diseased individuals often lack certain probiotic gene modules, with significant variation of the missing modules across different diseases. Additionally, different gene modules on the same probiotic have heterogeneous effects on various diseases. We thus believe that gene function integrity of the probiotic community is more crucial in maintaining gut homeostasis than merely increasing specific gene abundance, and adding probiotics indiscriminately might not boost health. We expect that the innovative language model-based metaProbiotics tool will promote novel probiotic discovery using large-scale metagenomic data and facilitate systematic research on bacterial probiotic effects. The metaProbiotics program can be freely downloaded at https://github.com/zhenchengfang/metaProbiotics.

High-throughput prediction of enzyme promiscuity based on substrate-product pairs

The screening of enzymes for catalyzing specific substrate-product pairs is often constrained in the realms of metabolic engineering and synthetic biology. Existing tools based on substrate and reaction similarity predominantly rely on prior knowledge, demonstrating limited extrapolative capabilities and an inability to incorporate custom candidate-enzyme libraries. Addressing these limitations, we have developed the Substrate-product Pair-based Enzyme Promiscuity Prediction (SPEPP) model. This innovative approach utilizes transfer learning and transformer architecture to predict enzyme promiscuity, thereby elucidating the intricate interplay between enzymes and substrate-product pairs. SPEPP exhibited robust predictive ability, eliminating the need for prior knowledge of reactions and allowing users to define their own candidate-enzyme libraries. It can be seamlessly integrated into various applications, including metabolic engineering, de novo pathway design, and hazardous material degradation. To better assist metabolic engineers in designing and refining biochemical pathways, particularly those without programming skills, we also designed EnzyPick, an easy-to-use web server for enzyme screening based on SPEPP. EnzyPick is accessible at http://www.biosynther.com/enzypick/.

Multi-modal features-based human-herpesvirus protein-protein interaction prediction by using LightGBM

The identification of human-herpesvirus protein-protein interactions (PPIs) is an essential and important entry point to understand the mechanisms of viral infection, especially in malignant tumor patients with common herpesvirus infection. While natural language processing (NLP)-based embedding techniques have emerged as powerful approaches, the application of multi-modal embedding feature fusion to predict human-herpesvirus PPIs is still limited. Here, we established a multi-modal embedding feature fusion-based LightGBM method to predict human-herpesvirus PPIs. In particular, we applied document and graph embedding approaches to represent sequence, network and function modal features of human and herpesviral proteins. Training our LightGBM models through our compiled non-rigorous and rigorous benchmarking datasets, we obtained significantly better performance compared to individual-modal features. Furthermore, our model outperformed traditional feature encodings-based machine learning methods and state-of-the-art deep learning-based methods using various benchmarking datasets. In a transfer learning step, we show that our model that was trained on human-herpesvirus PPI dataset without cytomegalovirus data can reliably predict human-cytomegalovirus PPIs, indicating that our method can comprehensively capture multi-modal fusion features of protein interactions across various herpesvirus subtypes. The implementation of our method is available at https://github.com/XiaodiYangpku/MultimodalPPI/.

MehNet: a vigesimal-based model by amino acid melting points generates unique ID numbers for protein sequences

The amino acid encoding plays a pivotal role in machine learning-based methods for predicting protein structure and function, as well as in protein mapping techniques. Additionally, the classification of protein sequences presents its own challenges. The current study aims to assign a constant value to each amino acid, thereby creating distinctions among protein sequences. The datasets used in this study were obtained from the UniProt Knowledgebase. Subsequently, these datasets underwent preprocessing steps, and identical sequences were categorized under the same headings. Each amino acid was ranked based on its respective melting point and was assigned a vigesimal digit. These generated vigesimal digits were subsequently converted to decimal values. The centerpiece of this methodology was the melting point hashing table, which was given the name 'MehNet'. Ultimately, each protein sequence was assigned a unique identification number. This approach successfully digitized protein sequences. Notably, experiments involving randomly distributed vigesimal digits for amino acids did not yield results as promising as those achieved with MehNet. The model's classification phase, which utilizes a k-nearest neighbors (kNN) classifier, demonstrates exceptional performance in miscellaneous viral sequences. It achieves high accuracy rates, with an overall accuracy of 99.75%. Notably, it achieves an outstanding accuracy of 99.92% for the Influenza C class, highlighting its ability to distinguish closely related viral sequences.Communicated by Ramaswamy H. Sarma.

MOBFinder: a tool for mobilization typing of plasmid metagenomic fragments based on a language model

BACKGROUND

Mobilization typing (MOB) is a classification scheme for plasmid genomes based on their relaxase gene. The host ranges of plasmids of different MOB categories are diverse, and MOB is crucial for investigating plasmid mobilization, especially the transmission of resistance genes and virulence factors. However, MOB typing of plasmid metagenomic data is challenging due to the highly fragmented characteristics of metagenomic contigs.

RESULTS

We developed MOBFinder, an 11-class classifier, for categorizing plasmid fragments into 10 MOB types and a nonmobilizable category. We first performed MOB typing to classify complete plasmid genomes according to relaxase information and then constructed an artificial benchmark dataset of plasmid metagenomic fragments (PMFs) from those complete plasmid genomes whose MOB types are well annotated. Next, based on natural language models, we used word vectors to characterize the PMFs. Several random forest classification models were trained and integrated to predict fragments of different lengths. Evaluating the tool using the benchmark dataset, we found that MOBFinder outperforms previous tools such as MOBscan and MOB-suite, with an overall accuracy approximately 59% higher than that of MOB-suite. Moreover, the balanced accuracy, harmonic mean, and F1-score reached up to 99% for some MOB types. When applied to a cohort of patients with type 2 diabetes (T2D), MOBFinder offered insights suggesting that the MOBF type plasmid, which is widely present in Escherichia and Klebsiella, and the MOBQ type plasmid might accelerate antibiotic resistance transmission in patients with T2D.

CONCLUSIONS

To the best of our knowledge, MOBFinder is the first tool for MOB typing of PMFs. The tool is freely available at https://github.com/FengTaoSMU/MOBFinder.

Seq-InSite: sequence supersedes structure for protein interaction site prediction

MOTIVATION

Proteins accomplish cellular functions by interacting with each other, which makes the prediction of interaction sites a fundamental problem. As experimental methods are expensive and time consuming, computational prediction of the interaction sites has been studied extensively. Structure-based programs are the most accurate, while the sequence-based ones are much more widely applicable, as the sequences available outnumber the structures by two orders of magnitude. Ideally, we would like a tool that has the quality of the former and the applicability of the latter.

RESULTS

We provide here the first solution that achieves these two goals. Our new sequence-based program, Seq-InSite, greatly surpasses the performance of sequence-based models, matching the quality of state-of-the-art structure-based predictors, thus effectively superseding the need for models requiring structure. The predictive power of Seq-InSite is illustrated using an analysis of evolutionary conservation for four protein sequences.

AVAILABILITY AND IMPLEMENTATION

Seq-InSite is freely available as a web server at http://seq-insite.csd.uwo.ca/ and as free source code, including trained models and all datasets used for training and testing, at https://github.com/lucian-ilie/Seq-InSite.

NCSP-PLM: An ensemble learning framework for predicting non-classical secreted proteins based on protein language models and deep learning

Non-classical secreted proteins (NCSPs) refer to a group of proteins that are located in the extracellular environment despite the absence of signal peptides and motifs. They usually play different roles in intercellular communication. Therefore, the accurate prediction of NCSPs is a critical step to understanding in depth their associated secretion mechanisms. Since the experimental recognition of NCSPs is often costly and time-consuming, computational methods are desired. In this study, we proposed an ensemble learning framework, termed NCSP-PLM, for the identification of NCSPs by extracting feature embeddings from pre-trained protein language models (PLMs) as input to several fine-tuned deep learning models. First, we compared the performance of nine PLM embeddings by training three neural networks: Multi-layer perceptron (MLP), attention mechanism and bidirectional long short-term memory network (BiLSTM) and selected the best network model for each PLM embedding. Then, four models were excluded due to their below-average accuracies, and the remaining five models were integrated to perform the prediction of NCSPs based on the weighted voting. Finally, the 5-fold cross validation and the independent test were conducted to evaluate the performance of NCSP-PLM on the benchmark datasets. Based on the same independent dataset, the sensitivity and specificity of NCSP-PLM were 91.18% and 97.06%, respectively. Particularly, the overall accuracy of our model achieved 94.12%, which was 7~16% higher than that of the existing state-of-the-art predictors. It indicated that NCSP-PLM could serve as a useful tool for the annotation of NCSPs.

Machine Learning Methods in Protein-Protein Docking

An exponential increase in the number of publications that address artificial intelligence (AI) usage in life sciences has been noticed in recent years, while new modeling techniques are constantly being reported. The potential of these methods is vast-from understanding fundamental cellular processes to discovering new drugs and breakthrough therapies. Computational studies of protein-protein interactions, crucial for understanding the operation of biological systems, are no exception in this field. However, despite the rapid development of technology and the progress in developing new approaches, many aspects remain challenging to solve, such as predicting conformational changes in proteins, or more "trivial" issues as high-quality data in huge quantities.Therefore, this chapter focuses on a short introduction to various AI approaches to study protein-protein interactions, followed by a description of the most up-to-date algorithms and programs used for this purpose. Yet, given the considerable pace of development in this hot area of computational science, at the time you read this chapter, the development of the algorithms described, or the emergence of new (and better) ones should come as no surprise.

Human-virus protein-protein interactions maps assist in revealing the pathogenesis of viral infection

Many significant viral infections have been recorded in human history, which have caused enormous negative impacts worldwide. Human-virus protein-protein interactions (PPIs) mediate viral infection and immune processes in the host. The identification, quantification, localization, and construction of human-virus PPIs maps are critical prerequisites for understanding the biophysical basis of the viral invasion process and characterising the framework for all protein functions. With the technological revolution and the introduction of artificial intelligence, the human-virus PPIs maps have been expanded rapidly in the past decade and shed light on solving complicated biomedical problems. However, there is still a lack of prospective insight into the field. In this work, we comprehensively review and compare the effectiveness, potential, and limitations of diverse approaches for constructing large-scale PPIs maps in human-virus, including experimental methods based on biophysics and biochemistry, databases of human-virus PPIs, computational methods based on artificial intelligence, and tools for visualising PPIs maps. The work aims to provide a toolbox for researchers, hoping to better assist in deciphering the relationship between humans and viruses.

LSTM-SAGDTA: Predicting Drug-target Binding Affinity with an Attention Graph Neural Network and LSTM Approach

INTRODUCTION

Drug development is a challenging and costly process, yet it plays a crucial role in improving healthcare outcomes. Drug development requires extensive research and testing to meet the demands for economic efficiency, cures, and pain relief.

METHODS

Drug development is a vital research area that necessitates innovation and collaboration to achieve significant breakthroughs. Computer-aided drug design provides a promising avenue for drug discovery and development by reducing costs and improving the efficiency of drug design and testing.

RESULTS

In this study, a novel model, namely LSTM-SAGDTA, capable of accurately predicting drug-target binding affinity, was developed. We employed SeqVec for characterizing the protein and utilized the graph neural networks to capture information on drug molecules. By introducing self-attentive graph pooling, the model achieved greater accuracy and efficiency in predicting drug-target binding affinity.

CONCLUSION

Moreover, LSTM-SAGDTA obtained superior accuracy over current state-of-the-art methods only by using less training time. The results of experiments suggest that this method represents a highprecision solution for the DTA predictor.

A comprehensive framework for advanced protein classification and function prediction using synergistic approaches: Integrating bispectral analysis, machine learning, and deep learning

Proteins are fundamental components of diverse cellular systems and play crucial roles in a variety of disease processes. Consequently, it is crucial to comprehend their structure, function, and intricate interconnections. Classifying proteins into families or groups with comparable structural and functional characteristics is a crucial aspect of this comprehension. This classification is crucial for evolutionary research, predicting protein function, and identifying potential therapeutic targets. Sequence alignment and structure-based alignment are frequently ineffective techniques for identifying protein families.This study addresses the need for a more efficient and accurate technique for feature extraction and protein classification. The research proposes a novel method that integrates bispectrum characteristics, deep learning techniques, and machine learning algorithms to overcome the limitations of conventional methods. The proposed method uses numbers to represent protein sequences, utilizes bispectrum analysis, uses different topologies for convolutional neural networks to pull out features, and chooses robust features to classify protein families. The goal is to outperform existing methods for identifying protein families, thereby enhancing classification metrics. The materials consist of numerous protein datasets, whereas the methods incorporate bispectrum characteristics and deep learning strategies. The results of this study demonstrate that the proposed method for identifying protein families is superior to conventional approaches. Significantly enhanced quality metrics demonstrated the efficacy of the combined bispectrum and deep learning approaches. These findings have the potential to advance the field of protein biology and facilitate pharmaceutical innovation. In conclusion, this study presents a novel method that employs bispectrum characteristics and deep learning techniques to improve the precision and efficiency of protein family identification. The demonstrated advancements in classification metrics demonstrate this method's applicability to numerous scientific disciplines. This furthers our understanding of protein function and its implications for disease and treatment.

A CNN based m5c RNA methylation predictor

Post-transcriptional modifications of RNA play a key role in performing a variety of biological processes, such as stability and immune tolerance, RNA splicing, protein translation and RNA degradation. One of these RNA modifications is m5c which participates in various cellular functions like RNA structural stability and translation efficiency, got popularity among biologists. By applying biological experiments to detect RNA m5c methylation sites would require much more efforts, time and money. Most of the researchers are using pre-processed RNA sequences of 41 nucleotides where the methylated cytosine is in the center. Therefore, it is possible that some of the information around these motif may have lost. The conventional methods are unable to process the RNA sequence directly due to high dimensionality and thus need optimized techniques for better features extraction. To handle the above challenges the goal of this study is to employ an end-to-end, 1D CNN based model to classify and interpret m5c methylated data sites. Moreover, our aim is to analyze the sequence in its full length where the methylated cytosine may not be in the center. The evaluation of the proposed architecture showed a promising results by outperforming state-of-the-art techniques in terms of sensitivity and accuracy. Our model achieve 96.70% sensitivity and 96.21% accuracy for 41 nucleotides sequences while 96.10% accuracy for full length sequences.

HostNet: improved sequence representation in deep neural networks for virus-host prediction

BACKGROUND

The escalation of viruses over the past decade has highlighted the need to determine their respective hosts, particularly for emerging ones that pose a potential menace to the welfare of both human and animal life. Yet, the traditional means of ascertaining the host range of viruses, which involves field surveillance and laboratory experiments, is a laborious and demanding undertaking. A computational tool with the capability to reliably predict host ranges for novel viruses can provide timely responses in the prevention and control of emerging infectious diseases. The intricate nature of viral-host prediction involves issues such as data imbalance and deficiency. Therefore, developing highly accurate computational tools capable of predicting virus-host associations is a challenging and pressing demand.

RESULTS

To overcome the challenges of virus-host prediction, we present HostNet, a deep learning framework that utilizes a Transformer-CNN-BiGRU architecture and two enhanced sequence representation modules. The first module, k-mer to vector, pre-trains a background vector representation of k-mers from a broad range of virus sequences to address the issue of data deficiency. The second module, an adaptive sliding window, truncates virus sequences of various lengths to create a uniform number of informative and distinct samples for each sequence to address the issue of data imbalance. We assess HostNet's performance on a benchmark dataset of "Rabies lyssavirus" and an in-house dataset of "Flavivirus". Our results show that HostNet surpasses the state-of-the-art deep learning-based method in host-prediction accuracies and F1 score. The enhanced sequence representation modules, significantly improve HostNet's training generalization, performance in challenging classes, and stability.

CONCLUSION

HostNet is a promising framework for predicting virus hosts from genomic sequences, addressing challenges posed by sparse and varying-length virus sequence data. Our results demonstrate its potential as a valuable tool for virus-host prediction in various biological contexts. Virus-host prediction based on genomic sequences using deep neural networks is a promising approach to identifying their potential hosts accurately and efficiently, with significant impacts on public health, disease prevention, and vaccine development.

In silico protein function prediction: the rise of machine learning-based approaches

Proteins function as integral actors in essential life processes, rendering the realm of protein research a fundamental domain that possesses the potential to propel advancements in pharmaceuticals and disease investigation. Within the context of protein research, an imperious demand arises to uncover protein functionalities and untangle intricate mechanistic underpinnings. Due to the exorbitant costs and limited throughput inherent in experimental investigations, computational models offer a promising alternative to accelerate protein function annotation. In recent years, protein pre-training models have exhibited noteworthy advancement across multiple prediction tasks. This advancement highlights a notable prospect for effectively tackling the intricate downstream task associated with protein function prediction. In this review, we elucidate the historical evolution and research paradigms of computational methods for predicting protein function. Subsequently, we summarize the progress in protein and molecule representation as well as feature extraction techniques. Furthermore, we assess the performance of machine learning-based algorithms across various objectives in protein function prediction, thereby offering a comprehensive perspective on the progress within this field.

GraphTar: applying word2vec and graph neural networks to miRNA target prediction

BACKGROUND

MicroRNAs (miRNAs) are short, non-coding RNA molecules that regulate gene expression by binding to specific mRNAs, inhibiting their translation. They play a critical role in regulating various biological processes and are implicated in many diseases, including cardiovascular, oncological, gastrointestinal diseases, and viral infections. Computational methods that can identify potential miRNA-mRNA interactions from raw data use one-dimensional miRNA-mRNA duplex representations and simple sequence encoding techniques, which may limit their performance.

RESULTS

We have developed GraphTar, a new target prediction method that uses a novel graph-based representation to reflect the spatial structure of the miRNA-mRNA duplex. Unlike existing approaches, we use the word2vec method to accurately encode RNA sequence information. In conjunction with the novel encoding method, we use a graph neural network classifier that can accurately predict miRNA-mRNA interactions based on graph representation learning. As part of a comparative study, we evaluate three different node embedding approaches within the GraphTar framework and compare them with other state-of-the-art target prediction methods. The results show that the proposed method achieves similar performance to the best methods in the field and outperforms them on one of the datasets.

CONCLUSIONS

In this study, a novel miRNA target prediction approach called GraphTar is introduced. Results show that GraphTar is as effective as existing methods and even outperforms them in some cases, opening new avenues for further research. However, the expansion of available datasets is critical for advancing the field towards real-world applications.

An algorithm for drug discovery based on deep learning with an example of developing a drug for the treatment of lung cancer

In this study, we present an algorithmic framework integrated within the created software platform tailored for the discovery of novel small-molecule anti-tumor agents. Our approach was exemplified in the context of combatting lung cancer. In the initial phase, target identification for therapeutic intervention was accomplished. Leveraging deep learning, we scrutinized gene expression profiles, focusing on those associated with adverse clinical outcomes in lung cancer patients. Augmenting this, generative adversarial neural (GAN) networks were employed to amass additional patient data. This effort yielded a subset of genes definitively linked to unfavorable prognoses. We further employed deep learning to delineate genes capable of discriminating between normal and tumor tissues based on expression patterns. The remaining genes were earmarked as potential targets for precision lung cancer therapy. Subsequently, a dedicated module was formulated to predict the interactions between inhibitors and proteins. To achieve this, protein amino acid sequences and chemical compound formulations engaged in protein interactions were encoded into vectorized representations. Additionally, a deep learning-based component was developed to forecast IC50 values through experimentation on cell lines. Virtual pre-clinical trials employing these inhibitors facilitated the selection of pertinent cell lines for subsequent laboratory assays. In summary, our study culminated in the derivation of several small-molecule formulas projected to bind selectively to specific proteins. This algorithmic platform holds promise in accelerating the identification and design of anti-tumor compounds, a critical pursuit in advancing targeted cancer therapies.

Deep Learning for Genomics: From Early Neural Nets to Modern Large Language Models

The data explosion driven by advancements in genomic research, such as high-throughput sequencing techniques, is constantly challenging conventional methods used in genomics. In parallel with the urgent demand for robust algorithms, deep learning has succeeded in various fields such as vision, speech, and text processing. Yet genomics entails unique challenges to deep learning, since we expect a superhuman intelligence that explores beyond our knowledge to interpret the genome from deep learning. A powerful deep learning model should rely on the insightful utilization of task-specific knowledge. In this paper, we briefly discuss the strengths of different deep learning models from a genomic perspective so as to fit each particular task with proper deep learning-based architecture, and we remark on practical considerations of developing deep learning architectures for genomics. We also provide a concise review of deep learning applications in various aspects of genomic research and point out current challenges and potential research directions for future genomics applications. We believe the collaborative use of ever-growing diverse data and the fast iteration of deep learning models will continue to contribute to the future of genomics.

Domain-PFP allows protein function prediction using function-aware domain embedding representations

Domains are functional and structural units of proteins that govern various biological functions performed by the proteins. Therefore, the characterization of domains in a protein can serve as a proper functional representation of proteins. Here, we employ a self-supervised protocol to derive functionally consistent representations for domains by learning domain-Gene Ontology (GO) co-occurrences and associations. The domain embeddings we constructed turned out to be effective in performing actual function prediction tasks. Extensive evaluations showed that protein representations using the domain embeddings are superior to those of large-scale protein language models in GO prediction tasks. Moreover, the new function prediction method built on the domain embeddings, named Domain-PFP, substantially outperformed the state-of-the-art function predictors. Additionally, Domain-PFP demonstrated competitive performance in the CAFA3 evaluation, achieving overall the best performance among the top teams that participated in the assessment.

CLCAP: Contrastive learning improves antigenicity prediction for influenza A virus using convolutional neural networks

Influenza viruses are detected year-round over the world and the viruses will usually circulate during fall and winter, causing the seasonal flu. The growing novel variants of influenza viruses pose a significant concern to public health annually. However, the rapid mutation of the influenza viruses makes it challenging to timely track their evolution. Therefore, a fast, low-cost, and precise method to predict the antigenic variant of influenza viruses could help vaccine development and prevent viral transmission. In this study, we propose a multi-channel convolutional neural network using contrastive learning to predict the antigenicity of influenza A viruses. An integrated dataset containing antigenic data and protein sequences was collected from various public resources and literature. The experimental results on three different influenza subtypes indicate our proposed model outperforms other traditional machine learning classifiers for antigenicity prediction. In addition, it also demonstrates superior performance over several state-of-the-art approaches, with 5.18 %, 7.03 % and 7.82 % increase in accuracy compared to the best results for H1N1, H3N2 and H5N1, respectively. The proposed framework is timely and effective in influenza antigenicity prediction and can be adapted to the study of other viruses.

Deep neural network for discovering metabolism-related biomarkers for lung adenocarcinoma

Introduction

Lung cancer is a major cause of illness and death worldwide. Lung adenocarcinoma (LUAD) is its most common subtype. Metabolite-mRNA interactions play a crucial role in cancer metabolism. Thus, metabolism-related mRNAs are potential targets for cancer therapy.

Methods

This study constructed a network of metabolite-mRNA interactions (MMIs) using four databases. We retrieved mRNAs from the Tumor Genome Atlas (TCGA)-LUAD cohort showing significant expressional changes between tumor and non-tumor tissues and identified metabolism-related differential expression (DE) mRNAs among the MMIs. Candidate mRNAs showing significant contributions to the deep neural network (DNN) model were mined. Using MMIs and the results of function analysis, we created a subnetwork comprising candidate mRNAs and metabolites.

Results

Finally, 10 biomarkers were obtained after survival analysis and validation. Their good prognostic value in LUAD was validated in independent datasets. Their effectiveness was confirmed in the TCGA and an independent Clinical Proteomic Tumor Analysis Consortium (CPTAC) dataset by comparison with traditional machine-learning models.

Conclusion

To summarize, 10 metabolism-related biomarkers were identified, and their prognostic value was confirmed successfully through the MMI network and the DNN model. Our strategy bears implications to pave the way for investigating metabolic biomarkers in other cancers.

GcForest-based compound-protein interaction prediction model and its application in discovering small-molecule drugs targeting CD47

Identifying compound-protein interaction plays a vital role in drug discovery. Artificial intelligence (AI), especially machine learning (ML) and deep learning (DL) algorithms, are playing increasingly important roles in compound-protein interaction (CPI) prediction. However, ML relies on learning from large sample data. And the CPI for specific target often has a small amount of data available. To overcome the dilemma, we propose a virtual screening model, in which word2vec is used as an embedding tool to generate low-dimensional vectors of SMILES of compounds and amino acid sequences of proteins, and the modified multi-grained cascade forest based gcForest is used as the classifier. This proposed method is capable of constructing a model from raw data, adjusting model complexity according to the scale of datasets, especially for small scale datasets, and is robust with few hyper-parameters and without over-fitting. We found that the proposed model is superior to other CPI prediction models and performs well on the constructed challenging dataset. We finally predicted 2 new inhibitors for clusters of differentiation 47(CD47) which has few known inhibitors. The IC50s of enzyme activities of these 2 new small molecular inhibitors targeting CD47-SIRPα interaction are 3.57 and 4.79 μM respectively. These results fully demonstrate the competence of this concise but efficient tool for CPI prediction.

Pmf-cpi: assessing drug selectivity with a pretrained multi-functional model for compound-protein interactions

Compound-protein interactions (CPI) play significant roles in drug development. To avoid side effects, it is also crucial to evaluate drug selectivity when binding to different targets. However, most selectivity prediction models are constructed for specific targets with limited data. In this study, we present a pretrained multi-functional model for compound-protein interaction prediction (PMF-CPI) and fine-tune it to assess drug selectivity. This model uses recurrent neural networks to process the protein embedding based on the pretrained language model TAPE, extracts molecular information from a graph encoder, and produces the output from dense layers. PMF-CPI obtained the best performance compared to outstanding approaches on both the binding affinity regression and CPI classification tasks. Meanwhile, we apply the model to analyzing drug selectivity after fine-tuning it on three datasets related to specific targets, including human cytochrome P450s. The study shows that PMF-CPI can accurately predict different drug affinities or opposite interactions toward similar targets, recognizing selective drugs for precise therapeutics.Kindly confirm if corresponding authors affiliations are identified correctly and amend if any.Yes, it is correct.

Deep learning in preclinical antibody drug discovery and development

Antibody drugs have become a key part of biotherapeutics. Patients suffering from various diseases have benefited from antibody therapies. However, its development process is rather long, expensive and risky. To speed up the process, reduce cost and improve success rate, artificial intelligence, especially deep learning methods, have been widely used in all aspects of preclinical antibody drug development, from library generation to hit identification, developability screening, lead selection and optimization. In this review, we systematically summarize antibody encodings, deep learning architectures and models used in preclinical antibody drug discovery and development. We also critically discuss challenges and opportunities, problems and possible solutions, current applications and future directions of deep learning in antibody drug development.

Accelerating antibody discovery and design with artificial intelligence: Recent advances and prospects

Therapeutic antibodies are the largest class of biotherapeutics and have been successful in treating human diseases. However, the design and discovery of antibody drugs remains challenging and time-consuming. Recently, artificial intelligence technology has had an incredible impact on antibody design and discovery, resulting in significant advances in antibody discovery, optimization, and developability. This review summarizes major machine learning (ML) methods and their applications for computational predictors of antibody structure and antigen interface/interaction, as well as the evaluation of antibody developability. Additionally, this review addresses the current status of ML-based therapeutic antibodies under preclinical and clinical phases. While many challenges remain, ML may offer a new therapeutic option for the future direction of fully computational antibody design.

DeepBSRPred: deep learning-based binding site residue prediction for proteins

MOTIVATION

Proteins-protein interactions (PPIs) are important to govern several cellular activities. Amino acid residues, which are located at the interface are known as the binding sites and the information about binding sites helps to understand the binding affinities and functions of protein-protein complexes.

RESULTS

We have developed a deep neural network-based method, DeepBSRPred, for predicting the binding sites using protein sequence information and predicted structures from AlphaFold2. Specific sequence and structure-based features include position-specific scoring matrix (PSSM), solvent accessible surface area, conservation score and amino acid properties, and residue depth, respectively. Our method predicted the binding sites with an average F1 score of 0.73 in a dataset of 1236 proteins. Further, we compared the performance with other existing methods in the literature using four benchmark datasets and our method outperformed those methods.

AVAILABILITY AND IMPLEMENTATION

The DeepBSRPred web server can be found at https://web.iitm.ac.in/bioinfo2/deepbsrpred/index.html , along with all datasets used in this study. The trained models, the DeepBSRPred standalone source code, and the feature computation pipeline are freely available at https://web.iitm.ac.in/bioinfo2/deepbsrpred/download.html .