Particle filter has been widely applied into many fields in recent years. Combined with the deficiency analysis of particle filter, an adaptive optimized strategy for particle filter is proposed. This adaptive optimiz...
详细信息
Particle filter has been widely applied into many fields in recent years. Combined with the deficiency analysis of particle filter, an adaptive optimized strategy for particle filter is proposed. This adaptive optimized strategy includes two parts. One is an improved hybrid proposal distribution with adaptive parameter optimization for particle filter. Based on the performance analysis of different proposal distribution, a hybrid proposal distribution with adaptive annealing parameter optimization is utilized to consider current information of the latest observed measurement. The other is an adaptive resampling strategy based on diversity guidance. An adaptive resampling step in particle filter is tuned based on two diversity measures and an improved partial stratified resampling strategy is presented based on the weights optimal idea. With the simulation program, the performance of the proposed strategy is evaluated and its validity is verified.
This study has been carried out to give some scientific reasons for genome annotation, shorten the annotating time, and improve the results of gene prediction. Taking the sequence of the 6th chromosome, which has more...
详细信息
This study has been carried out to give some scientific reasons for genome annotation, shorten the annotating time, and improve the results of gene prediction. Taking the sequence of the 6th chromosome, which has more length sequences than others, of Oryza sativa L. ssp. japonica cv. Nipponbare as analysis data in this research, the gene prediction of monocots module, rice, has been done by using Fgenesh ver. 2.0, and the predicting results have been explored particularly by bioinformatics methods. Results showed that the number of predicted genes for this chromosome was very close to the number of TIGR annotated genes. The majority of the predicted genes were multi-exon genes which had a percentage of 77.52. Length range was very big in the predicted genes. According to the significant match number, multi-exon genes can be predicted more veracity than single exon genes and the support can be reached up to 100% by TIGR annotation and up to 78% by cDNA. From the angle of predicted exons location of multi-exon genes, the internal exons and last exons had a high support of cDNA. The length of internal exons was relatively short in high （〉95% length, 〉78% similarity） cDNA and/or TIGR annotation support multi-exon genes, but the first exons and last exons were on the reverse. The majority of single exon genes which had more than 95% in length, and 78% in similarity support by cDNA and/or TIGR annotation was relatively short in length. From the angle of exon number, the majority of the multi-exon genes of high （〉 95% length, 〉 78% similarity） cDNA and/or TIGR annotation support had no more than 5 exon number. It was concluded that the rice gene prediction by Fgenesh was very good but needed modification manually to some extent according to cDNA support after aligning the predicting sequence of genes with cDNA database of rice.
暂无评论