PATENA: An algorithm for the design of protein linker sequences

EGUINOA I; SÁNCHEZ IE

Bahía Blanca

Congreso; Sexto congreso de la Asociación Argentina de Bioinformática y Biología Computacional; 2015

A2B2C

The successful construction of multidomain fusion proteins requires a linker sequence to covalently join the selected domains. Usual requirements for this sequence are to lack any interfering functional feature and the adoption of an extended conformation, allowing globular domains to move freely Natural linkers are not always flexible and cannot be considered inert. Furthermore, the diversity of sequences available in linker databases does not usually fill the properties required by the protein engineering process. Hence, a rational approach could aid linker design. Sequences are evaluated for structured regions using the algorithms IUPRED[5], ANCHOR[7], TANGO[4], PASTA[6] and WALTZ[8]. Putative functional sites are searched using BLAST[1] and sequence patterns in the ELM[3] and PROSITE[2] databases. Net charge and UV absorption can also be evaluated at the user?s re- quest. Undesired structure and functional features are mapped to each sequence position, and the total number of undesired features is calculated, resulting in a global score. PATENA can find suitable protein linkers in a short execution time. The set of designs that can be obtained from the same sequence shows high diversity