A deep learning approach for orphan gene identification in moso bamboo (Phyllostachys edulis) based on the CNN + Transformer model

Zhang, Xiaodan; Xuan, Jinxiang; Yao, Chensong; Gao, Qijuan; Wang, Lianglong; Jin, Xiu; Li, Shaowen

doi:10.1186/s12859-022-04702-1

Cited by 9 publications

(7 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the increase in interest in orphan genes, researchers are using various approaches to identify and understand these unique genes with specific biological significance. Recently a new deep learning model (CNN + Transformer) was used to identify orphan genes in moso bamboo [35]. Another recently published website, eggNOG 6.0 database, provides a large number of species with functional annotations that allows the comparison of genes and potentially further our understanding of orphan genes [36].…”

Section: Discussionmentioning

confidence: 99%

ORFanID: A Web-Based Search Engine for the Discovery and Identification of Orphan and Taxonomically Restricted Genes

Gunasekera

Raja

Hewapathirana

et al. 2022

Preprint

View full text Add to dashboard Cite

With the multiplicity of genomes sequenced today, it has been shown that significant percentages of genes in any given taxon do not possess orthologous sequences in other taxa. These sequences are typically designated as orphans/ORFans when found as singletons in one species only or taxonomically restricted genes (TRGs) when found at higher taxonomic ranks. Therefore, quantitative and collective studies of these genes are necessary for understanding their biological origins. Currently, orphan gene identifying software is limited, and those previously available are either not functional, are limited in their database search range, or are too complex algorithmically. Thus, an interested researcher studying orphan genes must harvest their data from many disparate sources. ORFanID is a graphical web-based search engine that efficiently finds orphan genes and TRGs at all taxonomic levels, from DNA or amino acid sequences in the entire NCBI database cluster and other large bioinformatics repositories. To our knowledge, this is the first program that allows the identification of orphan genes using both nucleotide and protein sequences in any species of interest. ORFanID identifies genes unique to any taxonomic rank, from species to a domain, using standard NCBI systematic classifiers. The software allows for user control of the NCBI database search parameters. The results of the search are provided in a spreadsheet as well as a graphical display. All the tables in the software are sortable by column, and results can be easily filtered with fuzzy search functionality. In addition, the visual presentation is expandable and collapsible by taxonomy.

show abstract

Section: Discussionmentioning

confidence: 99%

ORFanID: A Web-Based Search Engine for the Discovery and Identification of Orphan and Taxonomically Restricted Genes

Gunasekera

Raja

Hewapathirana

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Meanwhile, the research on convolutional neural networks had also taken a new direction in recent years, in which the Transformer model in the field of Neuro-Linguistic Programming (NLP) was modified and transplanted to convolutional neural networks. The Transformer model was also added to the network, in comparison to the normal convolutional neural networks, to make up for the convolutional model’s failure to build relationships between features, which led to the incapability of fully utilizing the contextual feature information and low sensitivity to the global information [ 36 , 37 , 38 ]. The Transformer model was originally proposed by Google [ 39 ] and applied to the field of natural language processing to solve the problem of establishing the long contextual dependencies using the recurrent neural networks.…”

Section: Related Workmentioning

confidence: 99%

Table Tennis Track Detection Based on Temporal Feature Multiplexing Network

Liu

et al. 2023

Sensors

View full text Add to dashboard Cite

Recording the trajectory of table tennis balls in real-time enables the analysis of the opponent’s attacking characteristics and weaknesses. The current analysis of the ball paths mainly relied on human viewing, which lacked certain theoretical data support. In order to solve the problem of the lack of objective data analysis in the research of table tennis competition, a target detection algorithm-based table tennis trajectory extraction network was proposed to record the trajectory of the table tennis movement in video. The network improved the feature reuse rate in order to achieve a lightweight network and enhance the detection accuracy. The core of the network was the “feature store & return” module, which could store the output of the current network layer and pass the features to the input of the network layer at the next moment to achieve efficient reuse of the features. In this module, the Transformer model was used to secondarily process the features, build the global association information, and enhance the feature richness of the feature map. According to the designed experiments, the detection accuracy of the network was 96.8% for table tennis and 89.1% for target localization. Moreover, the parameter size of the model was only 7.68 MB, and the detection frame rate could reach 634.19 FPS using the hardware for the tests. In summary, the network designed in this paper has the characteristics of both lightweight and high precision in table tennis detection, and the performance of the proposed model significantly outperforms that of the existing models.

show abstract

“…Even though we have not given sufficient attention to these genes, comparative genomic studies have estimated that OGs constitute at least 1% of the total genes in a genome, depending on the alignment rate and taxonomic level considered ( Khalturin et al, 2009 ; Arendsee et al, 2014 ; Prabh and Rödelsperger, 2016 ). Several studies have been carried out to characterize OGs in plants at the species level: Arabidopsis ( Li and Wurtele, 2014 ), sweet orange ( Xu et al, 2015 ), rice ( Guo et al, 2007 ), moso bamboo ( Zhang et al, 2022 ), and at the family level: Brassicaceae ( Donoghue et al, 2011 ) and Poaceae ( Campbell et al, 2007 ). However, there is limited information about the function of most of these OGs, as they lack recognizable domains and functional motifs.…”

Section: Introductionmentioning

confidence: 99%

Taxonomically Restricted Genes Are Associated With Responses to Biotic and Abiotic Stresses in Sugarcane (Saccharum spp.)

et al. 2022

View full text Add to dashboard Cite

Orphan genes (OGs) are protein-coding genes that are restricted to particular clades or species and lack homology with genes from other organisms, making their biological functions difficult to predict. OGs can rapidly originate and become functional; consequently, they may support rapid adaptation to environmental changes. Extensive spread of mobile elements and whole-genome duplication occurred in the Saccharum group, which may have contributed to the origin and diversification of OGs in the sugarcane genome. Here, we identified and characterized OGs in sugarcane, examined their expression profiles across tissues and genotypes, and investigated their regulation under varying conditions. We identified 319 OGs in the Saccharum spontaneum genome without detected homology to protein-coding genes in green plants, except those belonging to Saccharinae. Transcriptomic analysis revealed 288 sugarcane OGs with detectable expression levels in at least one tissue or genotype. We observed similar expression patterns of OGs in sugarcane genotypes originating from the closest geographical locations. We also observed tissue-specific expression of some OGs, possibly indicating a complex regulatory process for maintaining diverse functional activity of these genes across sugarcane tissues and genotypes. Sixty-six OGs were differentially expressed under stress conditions, especially cold and osmotic stresses. Gene co-expression network and functional enrichment analyses suggested that sugarcane OGs are involved in several biological mechanisms, including stimulus response and defence mechanisms. These findings provide a valuable genomic resource for sugarcane researchers, especially those interested in selecting stress-responsive genes.

show abstract

A deep learning approach for orphan gene identification in moso bamboo (Phyllostachys edulis) based on the CNN + Transformer model

Cited by 9 publications

References 50 publications

ORFanID: A Web-Based Search Engine for the Discovery and Identification of Orphan and Taxonomically Restricted Genes

ORFanID: A Web-Based Search Engine for the Discovery and Identification of Orphan and Taxonomically Restricted Genes

Table Tennis Track Detection Based on Temporal Feature Multiplexing Network

Taxonomically Restricted Genes Are Associated With Responses to Biotic and Abiotic Stresses in Sugarcane (Saccharum spp.)

Contact Info

Product

Resources

About