Eger, D. | bioRxiv Science

bioRxiv2019

Armadillo repeat-containing protein 1 is a dual localization protein associated with mitochondrial intermembrane space bridging complex

Cristae architecture is important for the function of mitochondria, the organelles that play the central role in many cellular processes. The mitochondrial contact site and cristae organizing system (MICOS) together with the sorting and assembly machinery (SAM) forms the mitochondrial intermembrane space bridging complex (MIB), a large protein complex present in mammalian mitochondria that partakes in the formation and maintenance of cristae. We report here a new subunit of the mammalian MICOS/MIB complex, an armadillo repeat-containing protein 1 (ArmC1). ArmC1 localizes both to cytosol and mitochondria, where it associates with the outer mitochondrial membrane through its carboxy-terminus. ArmC1 interacts with other constituents of the MICOS/MIB complex and its amounts are reduced upon MICOS/MIB complex depletion. Mitochondria lacking ArmC1 do not show defects in cristae structure, respiration or protein content, but appear fragmented and with reduced motility. ArmC1 represents therefore a peripheral MICOS/MIB component that appears to play a role in mitochondrial distribution in the cell.

cell biology↗

bioRxiv2019

Deep Learning on Chaos Game Representation for Proteins

Classification of protein sequences is one big task in bioinformatics and has many applications. Different machine learning methods exist and are applied on these problems, such as support vector machines (SVM), random forests (RF), and neural networks (NN). All of these methods have in common that protein sequences have to be made machine-readable and comparable in the first step, for which different encodings exist. These encodings are typically based on physical or chemical properties of the sequence. However, due to the outstanding performance of deep neural networks (DNN) on image recognition, we used frequency matrix chaos game representation (FCGR) for encoding of protein sequences into images. In this study, we compare the performance of SVMs, RFs, and DNNs, trained on FCGR encoded protein sequences. While the original chaos game representation (CGR) has been used mainly for genome sequence encoding and classification, we modified it to work also for protein sequences, resulting in n-flakes representation, an image with several icosagons.\n\nWe could show that all applied machine learning techniques (RF, SVM, and DNN) show promising results compared to the state-of-the-art methods on our benchmark datasets, with DNNs outperforming the other methods and that FCGR is a promising new encoding method for protein sequences.

bioinformatics↗