Identification And Function Analysis Of Phytophthora Infestans Pathogenic-related SRNA Based On Random Forest And Biclustering Method

Potato is one of the important food and cash crops,and its yield and quality have important economic value and status to agricultural production and people's diet.Late blight caused by Phytophthora infestans is one of the main diseases of potato production.This disease can cause a substantial reduction in potato production and seriously threaten potato yield and quality.Starting from the infection mechanism of pathogens on host plants,in order to formulate persistent and broad-spectrum control strategies for them,this is one of the hotspots in studying plant fungal diseases.Therefore,studying the pathogenic mechanism of P.infestans to prevent late blight of potato and reduce its incidence is of great significance to potato production and other crops to increase yield and income.Recent studies indicate that fungal sRNA may be a potential transboundary pathogenic factor,which provides a new idea for studying the pathogenic mechanism of fungi.Based on the experimental research and data foundation related to fungal and plant interactions,this research uses machine learning and a combined method of biclustering and protein interaction network for recognition,feature analysis,and functional exploration of P.infestans pathogenic sRNA factors.First,the sRNA sequencing data of P.infestans and sRNA sequencing data of potato leaves infected by P.infestans were collected,and 450 sRNAs related to the pathogenicity of P.infestans were obtained by designing screening conditions and data comparison method.Through the analysis and discussion of several machine learning models,a random forest was selected as the classification and prediction model of the pathogenic sRNA of P.infestans.Second,the important features of the pathogenic sRNA were identified according to the model.The results indicate that the acquired sRNAs associated with the pathogenic process share many standard features,such as the 19 th,23rd nucleotide preferred G or U,and the 7th,9th,17 th nucleotide prefers U,A,and C,respectively.It suggests that these sRNAs have preferences at these several locations.Subsequently,this study analyzed the function of these pathogenic sRNAs.In the work,2034 differentially expressed potato transcripts under P.infestans stress and the whole potato transcripts were collected.The target of sRNA was predicted by TAPIR and a screening method was designed.The enrichment analysis of the predicted targets shows that many sRNA targets have a similar function,and these targets have interactions.Such as Serine-threonine protein kinase and heat shock protein,they play important roles in plant response to stress.The pathway enrichment of targets indicates that these sRNAs affect the normal growth and resistance process of potatoes,suggesting that the identified 450 P.infestans sRNAs may be specific for host infection.In addition,in this study,a multi-level iterative clustering method based on the combination of biclustering and protein interaction network was also proposed,and 123 potential sRNA transboundary regulation modules were excavated.This provides a new idea for the functional analysis of the pathogenic sRNA of P.infestans.This research has laid a solid foundation for further research and further pathological analysis of pathogenic fungi.It provides a new analytical method for understanding the molecular pathogenesis of plant fungal pathogens.