NGS read classification using AI

Clinical metagenomics is a powerful diagnostic tool, as it offers an open view into all DNA in a patient’s sample. This allows the detection of pathogens that would slip through the cracks of classical specific assays. However, due to this unspecific nature of metagenomic sequencing, a huge amount of unspecific data is generated during the sequencing itself and the diagnosis only takes place at the data analysis stage where relevant sequences are filtered out. Typically, this is done by comparison to reference databases. While this approach has been optimized over the past years and works well to detect pathogens that are represented in the used databases, a common challenge in analysing a metagenomic patient sample arises when no pathogen sequences are found: How to determine whether truly no evidence of a pathogen is present in the data or whether the pathogen’s genome is simply absent from the database and the sequences in the dataset could thus not be classified? Here, we present a novel approach to this problem of detecting novel pathogens in metagenomic datasets by classifying the (segments of) proteins encoded by the sequences in the datasets. We train a neural network on the sequences of coding sequences, labeled by taxonomic domain, and use this neural network to predict the taxonomic classification of sequences that can not be classified by comparison to a reference database, thus facilitating the detection of potential novel pathogens.

Discovery of Simple Blood-Based Test to Predict Early Onset of Alzheimer’s Using Standard Clinical Mass Spectrometry Platforms

Despite the increasing number of individuals affected by Alzheimer’s disease (AD) every year, no effective therapy has been developed to treat this neurodegenerative disease yet. The current methods for AD diagnosis are effective for clinical confirmation of the disease only when symptoms become apparent, years after molecular damage started within the patients’ brains. As higher expression of a conformationally altered p53 has been correlated with AD, we developed a mass spectrometry-based method for highly sensitive, specific, and reproducible quantification of a p53 conformational variant in plasma samples of patients with known clinical outcome. In particular, we tested the prognostic performance of an AD-specific 2D3A8-immunoselected p53 peptide (AZ 284™) in different sets of individuals progressing from both cognitively unimpaired (CU) and mild cognitive impairment (MCI) patients progressing to AD dementia. Our data showed that quantitative analysis of AZ 284™ is a reliable tool for predicting AD progression up to 6 years prior to dementia onset with AUC >90%. Taken together, these results support the implementation of p53 conformational variant quantification as an affordable and powerful diagnostic tool for early, non-invasive AD diagnosis.

2020 MR Safety for Cardiac Devices: An Update for Radiologists

Magnetic resonance imaging (MRI) is a unique and powerful diagnostic tool that provides images without ionizing radiation and, at times, can be the only modality to properly assess and diagnose some pathologies. Although many patients will need an MRI in their lifetime, many of them are still being unjustly denied access to it due to what were once considered absolute contraindications, including MR nonconditional pacemakers and implantable cardioverter-defibrillators. However, there are a number of large studies that have recently demonstrated that MRI can safely be performed in these patients under certain conditions. In addition, there are an increasing number of novel cardiac devices implanted in patients who may require an MRI. Radiologists need to familiarize themselves with these devices, identify which patients with these devices can safely undergo MRI, and under which conditions. In this article, we will review the current literature on MR safety and cardiac devices, elaborate on how to safely image patients with cardiac devices, and share the expertise of our tertiary cardiac institute.