Predicting functions of maize proteins using graph convolutional network

Functional annotation and meta-analysis of maize transcriptomes reveal genes involved in biotic and abiotic stress

BACKGROUND

Environmental stress factors, such as biotic and abiotic stress, are becoming more common due to climate variability, significantly affecting global maize yield. Transcriptome profiling studies provide insights into the molecular mechanisms underlying stress response in maize, though the functions of many genes are still unknown. To enhance the functional annotation of maize-specific genes, MaizeGDB has outlined a data-driven approach with an emphasis on identifying genes and traits related to biotic and abiotic stress.

RESULTS

We mapped high-quality RNA-Seq expression reads from 24 different publicly available datasets (17 abiotic and seven biotic studies) generated from the B73 cultivar to the recent version of the reference genome B73 (B73v5) and deduced stress-related functional annotation of maize gene models. We conducted a robust meta-analysis of the transcriptome profiles from the datasets to identify maize loci responsive to stress, identifying 3,230 differentially expressed genes (DEGs): 2,555 DEGs regulated in response to abiotic stress, 408 DEGs regulated during biotic stress, and 267 common DEGs (co-DEGs) that overlap between abiotic and biotic stress. We discovered hub genes from network analyses, and among the hub genes of the co-DEGs we identified a putative NAC domain transcription factor superfamily protein (Zm00001eb369060) IDP275, which previously responded to herbivory and drought stress. IDP275 was up-regulated in our analysis in response to eight different abiotic and four different biotic stresses. A gene set enrichment and pathway analysis of hub genes of the co-DEGs revealed hormone-mediated signaling processes and phenylpropanoid biosynthesis pathways, respectively. Using phylostratigraphic analysis, we also demonstrated how abiotic and biotic stress genes differentially evolve to adapt to changing environments.

CONCLUSIONS

These results will help facilitate the functional annotation of multiple stress response gene models and annotation in maize. Data can be accessed and downloaded at the Maize Genetics and Genomics Database (MaizeGDB).

Comparative protein analysis of two maize genotypes with contrasting tolerance to low temperature

BACKGROUND

Low temperature (LT) stress is one of the major environmental stress factors affecting the growth and yield of maize (Zea mays L.). Hence, it is important to unravel the molecular mechanisms behind LT stress tolerance to improve molecular breeding in LT tolerant genotypes. In the present study, two maize genotypes viz. Gurez local from Kashmir Himalaya and tropical grown GM6, were dissected for their LT stress response in terms of accumulation of differentially regulated proteins (DRPs). Leaf proteome analysis at three-leaf stage of maize seedlings subjected to LT stress of 6 °C for a total of 12 h duration was performed using two dimensional gel electrophoresis (2D-PAGE) followed by subsequent identification of the proteins involved.

RESULTS

After MALDI-TOF (Matrix-assisted laser desorption/ionization-time of flight) and bioinformatics analysis, 19 proteins were successfully identified in Gurez local, while as 10 proteins were found to get successful identification in GM6. The interesting observations from the present investigation is the identification of three novel proteins viz. threonine dehydratase biosynthetic chloroplastic, thylakoidal processing peptidase 1 chloroplastic, and nodulin-like protein, whose role in abiotic stress tolerance, in general, and LT stress, in particular, has not been reported so far. It is important to highlight here that most of LT responsive proteins including the three novel proteins were identified from Gurez local only, owing to its exceptional LT tolerance. From the protein profiles, obtained in both genotypes immediately after LT stress perception, it was inferred that stress responsive protein accumulation and their expression fashion help the Gurez local in seedling establishment and withstand unfavorable conditions as compared to GM6. This was inferred from the findings of pathway enrichment analysis like regulation of seed growth, timing of floral transition, lipid glycosylation, and aspartate family amino acid catabolic processes, besides other key stress defense mechanisms. However, in GM6, metabolic pathways enriched were found to be involved in more general processes including cell cycle DNA replication and regulation of phenylpropanoid metabolism. Furthermore, majority of the qRT-PCR results of the selected proteins demonstrated positive correlation between protein levels and transcript abundance, thereby strengthening our findings.

CONCLUSIONS

In conclusion, our findings reported majority of the identified proteins in Gurez local exhibiting up-regulated pattern under LT stress as compared to GM6. Furthermore, three novel proteins induced by LT stress were found in Gurez local, requiring further functional validation. Therefore, our results offer more insights for elucidating the molecular networks mediating LT stress tolerance in maize.

GOGCN: Graph Convolutional Network on Gene Ontology for Functional Similarity Analysis of Genes

The measurement of gene functional similarity plays a critical role in numerous biological applications, such as gene clustering, the construction of gene similarity networks. However, most existing approaches still rely heavily on traditional computational strategies, which are not guaranteed to achieve satisfactory performance. In this study, we propose a novel computational approach called GOGCN to measure gene functional similarity by modeling the Gene Ontology (GO) through Graph Convolutional Network (GCN). GOGCN is a graph-based approach that performs sufficient representation learning for terms and relations in the GO graph. First, GOGCN employs the GCN-based knowledge graph embedding (KGE) model to learn vector representations (i.e., embeddings) for all entities (i.e., terms). Second, GOGCN calculates the semantic similarity between two terms based on their corresponding vector representations. Finally, GOGCN estimates gene functional similarity by making use of the pair-wise strategy. During the representation learning period, GOGCN promotes semantic interaction between terms through GCN, thereby capturing the rich structural information of the GO graph. Further experimental results on various datasets suggest that GOGCN is superior to the other state-of-the-art approaches, which shows its reliability and effectiveness.

A systematic review of state-of-the-art strategies for machine learning-based protein function prediction

New drug discovery is inseparable from the discovery of drug targets, and the vast majority of the known targets are proteins. At the same time, proteins are essential structural and functional elements of living cells necessary for the maintenance of all forms of life. Therefore, protein functions have become the focus of many pharmacological and biological studies. Traditional experimental techniques are no longer adequate for rapidly growing annotation of protein sequences, and approaches to protein function prediction using computational methods have emerged and flourished. A significant trend has been to use machine learning to achieve this goal. In this review, approaches to protein function prediction based on the sequence, structure, protein-protein interaction (PPI) networks, and fusion of multi-information sources are discussed. The current status of research on protein function prediction using machine learning is considered, and existing challenges and prominent breakthroughs are discussed to provide ideas and methods for future studies.

Leveraging class hierarchy for detecting missing annotations on hierarchical multi-label classification

With the development of new sequencing technologies, availability of genomic data has grown exponentially. Over the past decade, numerous studies have used genomic data to identify associations between genes and biological functions. While these studies have shown success in annotating genes with functions, they often assume that genes are completely annotated and fail to take into account that datasets are sparse and noisy. This work proposes a method to detect missing annotations in the context of hierarchical multi-label classification. More precisely, our method exploits the relations of functions, represented as a hierarchy, by computing probabilities based on the paths of functions in the hierarchy. By performing several experiments on a variety of rice (Oriza sativa Japonica), we showcase that the proposed method accurately detects missing annotations and yields superior results when compared to state-of-art methods from the literature.

Tissue Specificity Based Isoform Function Prediction

Alternative splicing enables a gene spliced into different isoforms and hence protein variants. Identifying individual functions of these isoforms help deciphering the functional diversity of proteins. Although much efforts have been made for automatic gene function prediction, few efforts have been moved toward computational isoform function prediction, mainly due to the unavailable (or scanty) functional annotations of isoforms. Existing efforts directly combine multiple RNA-seq datasets without account of the important tissue specificity of alternative splicing. To bridge this gap, we introduce a novel approach called TS-Isofun to predict the functions of isoforms by integrating multiple functional association networks with respect to tissue specificity. TS-Isofun first constructs tissue-specific isoform functional association networks using multiple RNA-seq datasets from tissue-wise. Next, TS-Isofun assigns weights to these networks and models the tissue specificity by selectively integrating them with adaptive weights. It then introduces a joint matrix factorization-based data fusion model to leverage the integrated network, gene-level data and functional annotations of genes to infer the functions of isoforms. To achieve coherent weight assignment and isoform function prediction, TS-Isofun jointly optimizes the weights of individual networks and the isoform function prediction in a unified objective function. Experimental results show that TS-Isofun significantly outperforms state-of-the-art methods and the account of tissue specificity contributes to more accurate isoform function prediction.

Suitability Evaluation of Crop Variety via Graph Neural Network

With the continuous growth of the global population, insufficient food production has become an urgent problem to be solved in most countries. At present, using artificial intelligence technology to improve suitability between land and crop varieties to increase crop yields has become a consensus among agricultural researchers. However, there are still many problems in existing works, such as limited crop phenotypic data and the poor performance of artificial intelligence models. In this regard, we take maize as an example to collect a large amount of environmental climate and crop phenotypic traits data at multiple experimental sites and construct an extensive dataset. Then, we introduce a graph neural network model to learn crop suitability evaluation and finally achieve a good evaluation effect. The evaluation results of the model can not only provide a reference for expert evaluation but also judge the suitability of the variety to other test trial sites according to the data of the current one, so as to guide future breeding experiments.

Meta-analysis of QTLs and candidate genes associated with seed germination in rice (Oryza sativa L.)

Seed germination is one of the critical stages of plant life, and many quantitative trait loci (QTLs) control this complex trait. Meta-analysis of QTLs is a powerful computational technique for estimating the most stable QTLs regardless of the population's genetic background. Besides, this analysis effectively narrows down the confidence interval (CI) to identify candidate genes (CGs) and marker development. In the current study, a comprehensive genome-wide meta-analysis was performed on QTLs associated with germination in rice. This analysis was conducted based on the data reported over the last two decades. In this case, various analyses were performed, including seed germination rate, plumule length, radicle length, germination percentage, coleoptile length, coleorhiza length, radicle fresh weight, germination potential, and germination index. A total of 67 QTLs were projected onto a reference map for these traits and then integrated into 32 meta-QTLs (MQTLs) to provide a genetic framework for seed germination. The average CI of MQTLs was considerably reduced from 15.125 to 8.73 cM compared to the initial QTLs. This situation identified 728 well-known functionally characterized genes and novel putative CGs for investigated traits. The fold change calculation demonstrated that 155 CGs had significant changes in expression analysis. In this case, 112 and 43 CGs were up-regulated and down-regulated during germination, respectively. This study provides an overview and compares genetic loci controlling traits related to seed germination in rice. The findings can bridge the gap between QTLs and CGs for seed germination.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12298-022-01232-1.

PANDA2: protein function prediction using graph neural networks

High-throughput sequencing technologies have generated massive protein sequences, but the annotations of protein sequences highly rely on the low-throughput and expensive biological experiments. Therefore, accurate and fast computational alternatives are needed to infer functional knowledge from protein sequences. The gene ontology (GO) directed acyclic graph (DAG) contains the hierarchical relationships between GO terms but is hard to be integrated into machine learning algorithms for functional predictions. We developed a deep learning system named PANDA2 to predict protein functions, which used the cutting-edge graph neural network to model the topology of the GO DAG and integrated the features generated by transformer protein language models. Compared with the top 10 methods in CAFA3, PANDA2 ranked first in cellular component ontology (CCO), tied first in biological process ontology (BPO) but had a higher coverage rate, and second in molecular function ontology (MFO). Compared with other recently-developed cutting-edge predictors DeepGOPlus, GOLabeler, and DeepText2GO, and benchmarked on another independent dataset, PANDA2 ranked first in CCO, first in BPO, and second in MFO. PANDA2 can be freely accessed from http://dna.cs.miami.edu/PANDA2/.

Structure and dynamics of photoreceptor sensory cilia

The rod and cone photoreceptor cells of the vertebrate retina have highly specialized structures that enable them to carry out their function of light detection over a broad range of illumination intensities with optimized spatial and temporal resolution. Most prominent are their unusually large sensory cilia, consisting of outer segments packed with photosensitive disc membranes, a connecting cilium with many features reminiscent of the primary cilium transition zone, and a pair of centrioles forming a basal body which serves as the platform upon which the ciliary axoneme is assembled. These structures form a highway through which an enormous flux of material moves on a daily basis to sustain the continual turnover of outer segment discs and the energetic demands of phototransduction. After decades of study, the details of the fine structure and distribution of molecular components of these structures are still incompletely understood, but recent advances in cellular imaging techniques and animal models of inherited ciliary defects are yielding important new insights. This knowledge informs our understanding both of the mechanisms of trafficking and assembly and of the pathophysiological mechanisms of human blinding ciliopathies.

DMIL-IsoFun: predicting isoform function using deep multi-instance learning

MOTIVATION

Alternative splicing creates the considerable proteomic diversity and complexity on relatively limited genome. Proteoforms translated from alternatively spliced isoforms of a gene actually execute the biological functions of this gene, which reflect the functional knowledge of genes at a finer granular level. Recently, some computational approaches have been proposed to differentiate isoform functions using sequence and expression data. However, their performance is far from being desirable, mainly due to the imbalance and lack of annotations at isoform-level, and the difficulty of modeling gene-isoform relations.

RESULT

We propose a deep multi-instance learning based framework (DMIL-IsoFun) to differentiate the functions of isoforms. DMIL-IsoFun firstly introduces a multi-instance learning convolution neural network trained with isoform sequences and gene-level annotations to extract the feature vectors and initialize the annotations of isoforms, and then uses a class-imbalance Graph Convolution Network to refine the annotations of individual isoforms based on the isoform co-expression network and extracted features. Extensive experimental results show that DMIL-IsoFun improves the Smin and Fmax of state-of-the-art solutions by at least 29.6% and 40.8%. The effectiveness of DMIL-IsoFun is further confirmed on a testbed of human multiple-isoform genes, and Maize isoforms related with photosynthesis.

AVAILABILITY

The code and data are available at http://www.sdu-idea.cn/codes.php?name=DMIL-Isofun.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

DeepGraphGO: graph neural network for large-scale, multispecies protein function prediction

Motivation

Automated function prediction (AFP) of proteins is a large-scale multi-label classification problem. Two limitations of most network-based methods for AFP are (i) a single model must be trained for each species and (ii) protein sequence information is totally ignored. These limitations cause weaker performance than sequence-based methods. Thus, the challenge is how to develop a powerful network-based method for AFP to overcome these limitations.

Results

We propose DeepGraphGO, an end-to-end, multispecies graph neural network-based method for AFP, which makes the most of both protein sequence and high-order protein network information. Our multispecies strategy allows one single model to be trained for all species, indicating a larger number of training samples than existing methods. Extensive experiments with a large-scale dataset show that DeepGraphGO outperforms a number of competing state-of-the-art methods significantly, including DeepGOPlus and three representative network-based methods: GeneMANIA, deepNF and clusDCA. We further confirm the effectiveness of our multispecies strategy and the advantage of DeepGraphGO over so-called difficult proteins. Finally, we integrate DeepGraphGO into the state-of-the-art ensemble method, NetGO, as a component and achieve a further performance improvement.

Availability and implementation

https://github.com/yourh/DeepGraphGO.

Supplementary information

Supplementary data are available at Bioinformatics online.

Gene Ontology Meta Annotator for Plants (GOMAP)

Annotating gene structures and functions to genome assemblies is necessary to make assembly resources useful for biological inference. Gene Ontology (GO) term assignment is the most used functional annotation system, and new methods for GO assignment have improved the quality of GO-based function predictions. The Gene Ontology Meta Annotator for Plants (GOMAP) is an optimized, high-throughput, and reproducible pipeline for genome-scale GO annotation of plants. We containerized GOMAP to increase portability and reproducibility and also optimized its performance for HPC environments. Here we report on the pipeline's availability and performance for annotating large, repetitive plant genomes and describe how GOMAP was used to annotate multiple maize genomes as a test case. Assessment shows that GOMAP expands and improves the number of genes annotated and annotations assigned per gene as well as the quality (based on [Formula: see text]) of GO assignments in maize. GOMAP has been deployed to annotate other species including wheat, rice, barley, cotton, and soy. Instructions and access to the GOMAP Singularity container are freely available online at https://bioinformapping.com/gomap/ . A list of annotated genomes and links to data is maintained at https://dill-picl.org/projects/gomap/ .