Simulation Of Insect Intron Evolution Using Randomly Generated And Mutated Base Sequences

Introns are nucleotide sequences that exist in genes and are widely distributed in eukaryotes.At present,the origin and evolutionary mechanism of introns is still unclear.In recent decades,after the first discovery of introns in protein-coding genes of eukaryotes,the origin and evolution of introns in eukaryotes have attracted extensive attention.Till today,there are still two opposing hypotheses,i.e.“introns-early theory”and“introns-late theory”,that differently describe the evolutionary mechanism of introns.The evolution of introns is closely related to the evolution of eukaryotic genomes.Understanding the origin and evolutionary mechanism of introns is crucial to understanding the evolution of eukaryotic genes.Facing introns from different organisms which are greatly varied in length and base sequences,conventional evolutionary analysis methods failed in exploring whether current introns are evolved from a longer ancestral sequence through gradual mutating and deleting bases,or from a shorter ancestral sequence through inserting and mutating bases.Existing sequence simulations are mostly used to deduce the evolution of amino acid coding sequences of genes,which is not suitable for the simulation of intron sequence evolution.Therefore,it is necessary to establish a new analytical method to explore the evolutionary mechanisms of intron sequences.In present study,two models,i.e.“mutation-and-deletion”（MD）and“mutation-and-insertion”（MI）,were designed to simulate the evolution of introns for studying the evolutionary mechanism of intron sequences in eukaryotic genes.The main results are as follows:（1）We designed and compiled MD and MI programs by using C++computer language.The MD program starts with a relatively long sequence and generates new sequences by deleting a certain length of DNA fragment after mutating a certain number of bases.While the MI program starts with a relatively short sequence and generates new sequences by inserting a certain length of DNA fragment after mutating a certain number of bases.Both programs have 5 adjustable parameters,namely,L_AS1（length of the first ancestor sequence）,L_AS8（length of the eighth ancestor sequence）,M₁（Number of bases mutated per 1 branch length）,L_I/D（length of bases inserted or deleted each time）,M_I/D（number of mutated bases each time）.（2）Eleven species of insects from different orders were taken as the research object.Their Daughterless gene intron sequences were used to construct ML phylogenetic tree after multiple sequence alignment,and five characteristic values of the eleven insect intron sequences were found as follows:L_MSA（length of multiple sequence alignment）=3665 bp,R（ratio of transition to transversion under T92+G parameter model）=1.54,D?（overall mean distance）=1.52,SED?（standard error of the overall mean distance）=0.16,and TS_ML（topology score of the constructed ML tree）=22.（3）According to the L₁₆（4*5）orthogonal table,different levels of 5 parameters in the two programs were designed,and 16 MD models and 16 MI models were established respectively for simulation of intron sequence evolution.Statistical analyses on characteristic values of sequences generated by each model indicated that,in MD models,the characteristic value L_MSA is affected significantly by model parameters L_AS1,L_AS8 and M₁;D?is affected significantly by M₁,L_I/D,L_AS1 and M_I/D;SED?is affected significantly by M₁,L_I/D and M_I/D;and TS_ML is affected significantly by M₁.In MI models,L_MSA is affected significantly by M₁ and L_AS1;R is influenced significantly by L_I/D/D and M₁;D?is affected significantly by L_AS1;and TS_ML is affected significantly by L_AS1,L_I/D and M₁.（4）According to influence curves of various parameters on characteristic values of the generated sequences,the model parameters were optimized and referenced to set up different parameter combinations to run MD model and MI model again to simulate sequence evolution.The results showed that MD model with proper parameter settings could generate sequences with characteristic values matchable to insect introns,while MI models with all attempted parameter settings could not generate sequences with characteristic values matchable to insect introns.The above results indicate that the insect intron sequences for this study should have evolved gradually from longer ancestral sequences through base mutation accompanied by fragment deletion,which is more consistent with the evolutionary mechanism described by the introns-early theory.In this study,we designed and complied new computer programs for exploring the evolutionary mechanism of intron sequences.The obtained results can be used to verify which of the“introns-early theory”and“introns-late theory”is more consistent with the evolutionary path of intron sequences.It is of great significance for further elucidation of evolutionary mechanism of intron sequences.