Artificial intelligence for medical image analysis in epilepsy.

Given improvements in computing power, artificial intelligence (AI) with deep learning has emerged as the state-of-the art method for the analysis of medical imaging data and will increasingly be used in the clinical setting. Recent work in epilepsy research has aimed to use AI methods to improve diagnosis, prognosis, and treatment, with the ultimate goal of developing highly accurate and reliable tools to aid clinical decision making. Here, we review how researchers are currently using AI methods in the analysis of neuroimaging data in epilepsy, focusing on challenges unique to each imaging modality with an emphasis on clinical significance. We further provide critical analyses of existing techniques and recommend areas for future work. We call for: (1) a multimodal approach that leverages the strengths of different modalities while compensating for their individual weaknesses, and (2) widespread implementation of generalizability testing of proposed models, a needed step before their introduction into clinical workflows. To achieve both goals, more collaborations among research groups and institutions in this field will be required.

Enhancing Open Modification Searches via a Combined Approach Facilitated by Ursgal.

The identification of peptide sequences and their post-translational modifications (PTMs) is a crucial step in the analysis of bottom-up proteomics data. The recent development of open modification search (OMS) engines allows virtually all PTMs to be searched for. This not only increases the number of spectra that can be matched to peptides but also greatly advances the understanding of the biological roles of PTMs through the identification, and the thereby facilitated quantification, of peptidoforms (peptide sequences and their potential PTMs). Whereas the benefits of combining results from multiple protein database search engines have been previously established, similar approaches for OMS results have been missing so far. Here we compare and combine results from three different OMS engines, demonstrating an increase in peptide spectrum matches of 8-18%. The unification of search results furthermore allows for the combined downstream processing of search results, including the mapping to potential PTMs. Finally, we test for the ability of OMS engines to identify glycosylated peptides. The implementation of these engines in the Python framework Ursgal facilitates the straightforward application of the OMS with unified parameters and results files, thereby enabling yet unmatched high-throughput, large-scale data analysis.