Font Size: a A A

Research On Phenotype Screening Of Arabidopsis Thaliana By Heavy-ion Beams Based On High-throughput Imaging Technique

Posted on:2018-08-24Degree:MasterType:Thesis
Country:ChinaCandidate:J H MuFull Text:PDF
GTID:2310330533460038Subject:Biophysics
Abstract/Summary:PDF Full Text Request
In recent years,with the development of robotics,sensors and kinds of imaging apparatuses,scientists have developed varieties of automation,high precision and throughput phenotype analysis platforms by ceaselessly hard working and exploring.With all these high-throughput imaging system,high-precise sensor system,full-automatic robot assistant and strong data analytical processing software,the system platforms are available with phenotype acquisition as well as data analyzing and characteristics of high flux,non-destructive,and automatic,examining and analyzing root characters,yield,biomass,and stress responses.Although these platforms have improved the development of phenomics to some degree,there are still some problems in need of resolution.On the one hand,whether the high-flux phenotypic screening platform really screens mutants productively,researchers have not yet carried out an effective evaluation.On the other hand,only a few scientists are attempting to build an evaluation system based on the high throughout imaging system of corresponding plant phenotype and genotype.Certainly,it is the key to solve these problems that how to analyze thoroughly and availably the "big data" obtained from the "high throughput" of the phenotypic analysis platform.Therefore,based on these issues,we developed a comparative experiment of human-computer recognition.Two methods traditional snaked-eye observation and high-throughput phenotype observation system were applied to screen the mutants of Arabidopsis mutant and supplementary materials respectively.The snaked-eye observation adopted traditional approaches,while the machine measurement group analyzed huge amounts of data produced by Scanalyzer HTS using statistics analysis such as PCA,scatter matrix,and growth curve.And we compared the accuracy,and rate of false positive and false negative of the two methods finally.An effective mutant screening system was built eventually through integrating the exiting mutant information and man-computer experimental results.After the human-machine experiment,we first opened up an analysis route of phenomics with large data processing and mastered an useful mutant screening system.It is necessaryto reduce the dimension of big proteomics data obtained by HTS and compress the data amount at first.Then based on logistic function of plant growth curve,a synthetic method of principal component analysis(PCA)and scatter matrix clustering analysis is presented to screen plant variation.The system can screen more than 80% of the mutants from many plants.The experimental results show that(1)with the increase of the level of confidence interval in clustering analysis(from 95% to 99.9%).The accuracy rate of the machine screen mutants decreased,the false positive rate decreased gradually and false negative rate gradually increased.(2)within the 95%confidence interval,the machine accuracy is higher than that of the naked eye screening,and the false negative rate was lower than the naked eye screening,which means the machine can replace the human mutant screening to a certain extent,compared with people less prone to missing mutants.The false positive rate of the machine the value of screening in different confidence intervals were higher than that of the naked eye,compared with the time-consuming eye screening,use the machine for screening of mutants can achieve mass mutant screening,and can reduce the range of mutants in a certain range,improve the mutant screening efficiency.Hence it is possible to use machine screening to eliminate most of the non mutant plants and then use the naked eye to screen the mutants.(4)In our study we found that the accuracy of identifying all the site mutants was 16.7%,while the machine was 50% for same type of mutant grown in different locations.However,only 50% of the probability of the naked eye can only be identified in one of the two locations,and the machine's recognition probability is 16.7%.The results showed that compared with the traditional screening method with a certain subjectivity,the screening of the machine is more accurate,and the accuracy of the screening mutant is higher and the ability is stronger.
Keywords/Search Tags:Carbon ion beam irradiation, Arabidopsis thaliana, Mutation screening, High throughput, Phenomics
PDF Full Text Request
Related items