In the last decade, next generation sequencing technologies have made a great progress, allowing the generation of a huge amount of sequence data with just a tiny portion of the cost and effort that was required with traditional technologies. These advances have transformed biomedical research, specially in cancer in where the use of these technologies has unraveled very interesting and complex landscapes of somatic variants. However, when the advances in next-generation sequencing have been huge, this has come with the cost of an amazing increase of the computational load involved in analysis of this data. Although several algorithms to call mutations have been developed in these years, they encounter with great diffi culties in the analysis of cancer sequencing data. In particular, in this type of samples, some mutations are harbored by a low frequency of the cells as a consequence of the presence of normal DNA contamination or intratumor heterogeneity. These mutations present in minority clones are diffi cult to distinguish from sequencing and mapping errors. Considering this background, we have decided to set up a new system specially designed to take into account all these diffi culties in cancer samples in order to obtain the maximum sensitivity on the calling without sacrifi cing specificity. Additionally we have performed a benchmark comparison of several softwares available with a our in house written software that we have called RAMSES (Realignment Assisted Minimum Somatic Spotter). Together with a good performance compared with other available software, RAMSES offers a great flexibility for the user, counts with filters to recently described bias like the oxidative DNA damage produced during library preparation, and incorporates new functionalities difficult to find in other softwares. This includes calling of both substitutions and small insertions and deletions (indels) with the help of PINDEL realigner and the automatic functional annotation of the mutations found using Ensembl transcript database.

Resumen del trabajo presentado al XXXXVIII Congreso de la Sociedad Española de Bioquímica y Biología Molecular (SEBBM), celebrado en Valencia del 7 al 10 de septiembre de 2015.

Peer reviewed