Predicting the gene structure due to alternative splicing in the NGS era

Large-scale genomic and trascriptomic analysis in eukaryotic genome projects have produced   surprising results that have strongly challenged the traditional understanding of genes.  In particular, the gene-centric
view of molecular biologyhas been changed in favour of considering function products (proteins and ncRNAs) rather than genes. This novel view is a consequence  of some important findings obtained by a widespread use of software tools specifically developed to accomplish the important  task of predicting the gene structure from large-scale gene products. In this talk, we will discuss  this task by presenting some challenging computational problems and  their algorithmic solutions.  We will show that  some open problems involve the  solution of recurrent computational issues related to the use of next generation sequencing data
(NGS) such as for example de novo assembly, indexing and graph representation of large collections of strings.