A Preliminary Comparison of P-Tool Consistency Javier Murillo 1(B ) , Flavio Spetale 1 , Elizabeth Tapia 1 , Flavia Krsticevic 6 , Olivier Cailloux 2 , Serge Guillaume 3 , Gustavo Vazquez 4 , Tamara Fernandez 4 , Sebastien Destercke 5 , Sergio Ponce 6 , and Pilar Bulacio 1 1 CIFASIS-CONICET, Univ. Nacional de Rosario, Rosario, Argentina murillo@cifasis-conicet.gov.ar 2 Universit´ e Paris-Dauphine, CNRS, Paris, France 3 ITAP, Irstea, Montpellier SupAgro, Univ. Montpellier, Montpellier, France 4 Universidad Cat´olica del Uruguay, Montevideo, Uruguay 5 Universit´ e de Technologie de Compi` egne, Compiegne, France 6 Universidad Tecnol´ogica Nacional, Regional San Nicol´as, Buenos Aires, Argentina Abstract. Many Bioinformatics tools, known as p-tools, have been developed to predict the effect of single nucleotide polymorphisms (SNPs) on gene functionality, in an effort to reduce the need for in-vivo assays. However, the large number of p-tools available and the hetero- geneity of their output make their selection and comparison difficult. To study the consistency of predictions across p-tools, here we present two indices and test them on five p-tools whose predictions are based on different types of background information. For this test, SNPs from well-known organism Drosophila melanogaster are considered. Keywords: SNP · Gene functionality · Missense nonsense mutation 1 Introduction A main factor underlying the conformation of proteins is their amino acid sequence. An individual nucleotide change, also called a Single Nucleotide Poly- morphism (SNP), is a missense mutation when it causes a different protein, or a nonsense mutation when it causes a short and non-functional protein. The degree to which a SNP affects protein function is a key point, but its prediction remains an open problem. Next-Generation Sequencing (NGS) technologies have made it possible to detect thousands of SNPs [1], but wet-lab studies needed to associate these SNPs with phenotypic traits are costly. To narrow down the list of candidate SNPs, several Bioinformatics tools, hereafter referred to as p-tools, have been developed to predict the impact of SNPs in-silico. P-tools can be based on infor- mation from amino acid sequences, protein structure, context, functional param- eters and evolutionary information [2]. For instance, for sequence conservation c  Springer Nature Switzerland AG 2020 C. A. Gonz´alez D´ ıaz et al. (Eds.): CLAIB 2019, IFMBE Proceedings 75, pp. 731–735, 2020. https://doi.org/10.1007/978-3-030-30648-9_97