Algorithmic Approach for finding Convolutional Code generators for the Translation Initiation of Escherichia coli K-12.

Abstract

Using error-control coding theory, we parallel the functionality of the translation of mRNA into amino acids to the decoding of noisy parity streams that have been encoded using a convolutional code. This enables us to model the ribosome as a table-based convolution decoder. In this work, we attempt to find plausible convolutional code generators for the translation initiation of Escherichia coli K-12. We choose the g-mask from the exposed part of the 16S rRNA. We develop an algorithmic approach to calculate the generators from the g-mask. We assign plausibility to the generators based on their ability to produce encoded sequences which exhibit a clear distinction between the translated and non-translated sequences. We also explore the construction of g-masks based on binding patterns, and evaluate the performance of the corresponding generators.