Single-cell transcriptomics is a powerful approach for characterizing gene transcription at cellular resolution. This approach requires efficient computational pipelines to undertake essential tasks, including clustering, dimensionality reduction, imputation, and denoising. Currently, most such pipelines undertake these computational tasks separately without considering the interdependence among these tasks. Here, we present an advanced pipeline, MUSIC-GCN, by employing a graph convolutional neural (GCN) network and autoencoder to perform multi-task single-cell RNA-sequencing (scRNA-seq) data analysis. The rationale is that multiple related tasks can be carried out simultaneously to enable enhanced learning and more effective representations through the 'sharing of knowledge' regarding individual tasks. Benchmarking experiments using various scRNA-seq datasets show that MUSIC-GCN can achieve a competitive performance on multi-tasks when benchmarked with state-of-the-art approaches.

The identification and characterization of essential genes are central to our understanding of the core biological functions in eukaryotic organisms, and has important implications for the treatment of diseases caused by, for example, cancers and pathogens. Given the major constraints in testing the functions of genes of many organisms in the laboratory, due to the absence of in vitro cultures and/or gene perturbation assays for most metazoan species, there has been a need to develop in silico tools for the accurate prediction or inference of essential genes to underpin systems biological investigations. Major advances in machine learning approaches provide unprecedented opportunities to overcome these limitations and accelerate the discovery of essential genes on a genome-wide scale. Here, we developed and evaluated a large language model- and graph neural network (LLM-GNN)-based approach, called 'Bingo', to predict essential protein-coding genes in the metazoan model organisms Caenorhabditis elegans and Drosophila melanogaster as well as in Mus musculus and Homo sapiens (a HepG2 cell line) by integrating LLM and GNNs with adversarial training. Bingo predicts essential genes under two 'zero-shot' scenarios with transfer learning, showing promise to compensate for a lack of high-quality genomic and proteomic data for non-model organisms. In addition, the attention mechanisms and GNNExplainer were employed to manifest the functional sites and structural domain with most contribution to essentiality. In conclusion, Bingo provides the prospect of being able to accurately infer the essential genes of little- or under-studied organisms of interest, and provides a biological explanation for gene essentiality.

Single-cell RNA sequencing (scRNA-seq) has significantly accelerated the experimental characterization of distinct cell lineages and types in complex tissues and organisms. Cell-type annotation is of great importance in most of the scRNA-seq analysis pipelines. However, manual cell-type annotation heavily relies on the quality of scRNA-seq data and marker genes, and therefore can be laborious and time-consuming. Furthermore, the heterogeneity of scRNA-seq datasets poses another challenge for accurate cell-type annotation, such as the batch effect induced by different scRNA-seq protocols and samples. To overcome these limitations, here we propose a novel pipeline, termed TripletCell, for cross-species, cross-protocol and cross-sample cell-type annotation. We developed a cell embedding and dimension-reduction module for the feature extraction (FE) in TripletCell, namely TripletCell-FE, to leverage the deep metric learning-based algorithm for the relationships between the reference gene expression matrix and the query cells. Our experimental studies on 21 datasets (covering nine scRNA-seq protocols, two species and three tissues) demonstrate that TripletCell outperformed state-of-the-art approaches for cell-type annotation. More importantly, regardless of protocols or species, TripletCell can deliver outstanding and robust performance in annotating different types of cells. TripletCell is freely available at https://github.com/liuyan3056/TripletCell. We believe that TripletCell is a reliable computational tool for accurately annotating various cell types using scRNA-seq data and will be instrumental in assisting the generation of novel biological hypotheses in cell biology.

MOTIVATION: The rapid accumulation of high-throughput sequence data demands the development of effective and efficient data-driven computational methods to functionally annotate proteins. However, most current approaches used for functional annotation simply focus on the use of protein-level information but ignore inter-relationships among annotations. RESULTS: Here, we established PFresGO, an attention-based deep-learning approach that incorporates hierarchical structures in Gene Ontology (GO) graphs and advances in natural language processing algorithms for the functional annotation of proteins. PFresGO employs a self-attention operation to capture the inter-relationships of GO terms, updates its embedding accordingly and uses a cross-attention operation to project protein representations and GO embedding into a common latent space to identify global protein sequence patterns and local functional residues. We demonstrate that PFresGO consistently achieves superior performance across GO categories when compared with 'state-of-the-art' methods. Importantly, we show that PFresGO can identify functionally important residues in protein sequences by assessing the distribution of attention weightings. PFresGO should serve as an effective tool for the accurate functional annotation of proteins and functional domains within proteins. AVAILABILITY AND IMPLEMENTATION: PFresGO is available for academic purposes at https://github.com/BioColLab/PFresGO. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The molecular subtyping of gastric cancer (adenocarcinoma) into four main subtypes based on integrated multiomics profiles, as proposed by The Cancer Genome Atlas (TCGA) initiative, represents an effective strategy for patient stratification. However, this approach requires the use of multiple technological platforms, and is quite expensive and time-consuming to perform. A computational approach that uses histopathological image data to infer molecular subtypes could be a practical, cost- and time-efficient complementary tool for prognostic and clinical management purposes. RESULTS: Here, we propose a deep learning ensemble approach (called DEMoS) capable of predicting the four recognized molecular subtypes of gastric cancer directly from histopathological images. DEMoS achieved tile-level area under the receiver-operating characteristic curve (AUROC) values of 0.785, 0.668, 0.762 and 0.811 for the prediction of these four subtypes of gastric cancer [i.e. (i) Epstein-Barr (EBV)-infected, (ii) microsatellite instability (MSI), (iii) genomically stable (GS) and (iv) chromosomally unstable tumors (CIN)] using an independent test dataset, respectively. At the patient-level, it achieved AUROC values of 0.897, 0.764, 0.890 and 0.898, respectively. Thus, these four subtypes are well-predicted by DEMoS. Benchmarking experiments further suggest that DEMoS is able to achieve an improved classification performance for image-based subtyping and prevent model overfitting. This study highlights the feasibility of using a deep learning ensemble-based method to rapidly and reliably subtype gastric cancer (adenocarcinoma) solely using features from histopathological images. AVAILABILITY AND IMPLEMENTATION: All whole slide images used in this study was collected from the TCGA database. This study builds upon our previously published HEAL framework, with related documentation and tutorials available at http://heal.erc.monash.edu.au. The source code and related models are freely accessible at https://github.com/Docurdt/DEMoS.git. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Analyses of replicates in sets of discrete data, typically acquired in multi-well plate formats, is a recurring task in many contemporary areas in the Life Sciences. The availability of accessible cross-platform data analysis tools for such fundamental tasks in varied projects and environments is an important prerequisite to ensuring a reliable and timely turnaround as well as to provide practical analytical tools for student training. RESULTS: We have developed an easy-to-use, interactive software tool for the analysis of multiple data sets comprising replicates of discrete bivariate data points. For each dataset, the software identifies the replicate data points from a defined matrix layout and calculates their means and standard errors. The averaged values are then automatically fitted using either a linear or a logistic dose response function. CONCLUSIONS: DRfit is a practical and convenient tool for the analysis of one or multiple sets of discrete data points acquired as replicates from multi-well plate assays. The design of the graphical user interface and the built-in analysis features make it a flexible and useful tool for a wide range of different assays.

SUMMARY: Both alignment generation and visualization are important processes for producing biologically meaningful sequence alignments. Computational tools that combine reliable, automated and semi-automated approaches to produce secondary structure-based alignments with an appropriate visualization of the results are rare. We have developed SBAL, a tool to generate and edit secondary structure-based sequence alignments. It is easy to install and provides a user-friendly interface. Sequence alignments are displayed, with secondary structure assignments mapped to their corresponding regions in the sequence by using a simple colour scheme. The algorithm implemented for automated and semi-automated secondary structure-based alignment calculations shows a comparable performance to existing software. AVAILABILITY AND IMPLEMENTATION: SBAL has been implemented in Java to provide cross-platform compatibility. SBAL is freely available to academic users at http://www.structuralchemistry.org/pcsb/. Users will be asked for their name, institution and email address. A manual can also be downloaded from this site. The software, manual and test sets are also available as supplementary material. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: The transcriptome of an organism can be studied with the analysis of expressed sequence tag (EST) data sets that offers a rapid and cost effective approach with several new and updated bioinformatics approaches and tools for assembly and annotation. The comprehensive analyses comprehend an organism along with the genome and proteome analysis. With the advent of large-scale sequencing projects and generation of sequence data at protein and cDNA levels, automated analysis pipeline is necessary to store, organize and annotate ESTs. RESULTS: TranSeqAnnotator is a workflow for large-scale analysis of transcriptomic data with the most appropriate bioinformatics tools for data management and analysis. The pipeline automatically cleans, clusters, assembles and generates consensus sequences, conceptually translates these into possible protein products and assigns putative function based on various DNA and protein similarity searches. Excretory/secretory (ES) proteins inferred from ESTs/short reads are also identified. The TranSeqAnnotator accepts FASTA format raw and quality ESTs along with protein and short read sequences and are analysed with user selected programs. After pre-processing and assembly, the dataset is annotated at the nucleotide, protein and ES protein levels. CONCLUSION: TranSeqAnnotator has been developed in a Linux cluster, to perform an exhaustive and reliable analysis and provide detailed annotation. TranSeqAnnotator outputs gene ontologies, protein functional identifications in terms of mapping to protein domains and metabolic pathways. The pipeline is applied to annotate large EST datasets to identify several novel and known genes with therapeutic experimental validations and could serve as potential targets for parasite intervention. TransSeqAnnotator is freely available for the scientific community at http://estexplorer.biolinfo.org/TranSeqAnnotator/.

BACKGROUND: The analysis of expressed sequence tags (EST) offers a rapid and cost effective approach to elucidate the transcriptome of an organism, but requires several computational methods for assembly and annotation. Researchers frequently analyse each step manually, which is laborious and time consuming. We have recently developed ESTExplorer, a semi-automated computational workflow system, in order to achieve the rapid analysis of EST datasets. In this study, we evaluated EST data analysis for the parasitic nematode Trichostrongylus vitrinus (order Strongylida) using ESTExplorer, compared with database matching alone. RESULTS: We functionally annotated 1776 ESTs obtained via suppressive-subtractive hybridisation from T. vitrinus, an important parasitic trichostrongylid of small ruminants. Cluster and comparative genomic analyses of the transcripts using ESTExplorer indicated that 290 (41%) sequences had homologues in Caenorhabditis elegans, 329 (42%) in parasitic nematodes, 202 (28%) in organisms other than nematodes, and 218 (31%) had no significant match to any sequence in the current databases. Of the C. elegans homologues, 90 were associated with 'non-wildtype' double-stranded RNA interference (RNAi) phenotypes, including embryonic lethality, maternal sterility, sterile progeny, larval arrest and slow growth. We could functionally classify 267 (38%) sequences using the Gene Ontologies (GO) and establish pathway associations for 230 (33%) sequences using the Kyoto Encyclopedia of Genes and Genomes (KEGG). Further examination of this EST dataset revealed a number of signalling molecules, proteases, protease inhibitors, enzymes, ion channels and immune-related genes. In addition, we identified 40 putative secreted proteins that could represent potential candidates for developing novel anthelmintics or vaccines. We further compared the automated EST sequence annotations, using ESTExplorer, with database search results for individual T. vitrinus ESTs. ESTExplorer reliably and rapidly annotated 301 ESTs, with pathway and GO information, eliminating 60 low quality hits from database searches. CONCLUSION: We evaluated the efficacy of ESTExplorer in analysing EST data, and demonstrate that computational tools can be used to accelerate the process of gene discovery in EST sequencing projects. The present study has elucidated sets of relatively conserved and potentially novel genes for biological investigation, and the annotated EST set provides further insight into the molecular biology of T. vitrinus, towards the identification of novel drug targets.

Expressed sequence tag (EST) sequencing projects are underway for numerous organisms, generating millions of short, single-pass nucleotide sequence reads, accumulating in EST databases. Extensive computational strategies have been developed to organize and analyse both small- and large-scale EST data for gene discovery, transcript and single nucleotide polymorphism analysis as well as functional annotation of putative gene products. We provide an overview of the significance of ESTs in the genomic era, their properties and the applications of ESTs. Methods adopted for each step of EST analysis by various research groups have been compared. Challenges that lie ahead in organizing and analysing the ever increasing EST data have also been identified. The most appropriate software tools for EST pre-processing, clustering and assembly, database matching and functional annotation have been compiled (available online from http://biolinfo.org/EST). We propose a road map for EST analysis to accelerate the effective analyses of EST data sets. An investigation of EST analysis platforms reveals that they all terminate prior to downstream functional annotation including gene ontologies, motif/pattern analysis and pathway mapping.