Genomic surveillance of the SARS-CoV-2 pandemic is crucial and mainly achieved by amplicon sequencing protocols. Overlapping tiled-amplicons are generated to establish contiguous SARS-CoV-2 genome sequences, which enable the precise resolution of infection chains and outbreaks. We investigated a SARS-CoV-2 outbreak in a local hospital and used nanopore sequencing with a modified ARTIC protocol employing 1200 bp long amplicons. We detected a long deletion of 168 nucleotides in the ORF8 gene in 76 samples from the hospital outbreak. This deletion is difficult to identify with the classical amplicon sequencing procedures since it removes two amplicon primer-binding sites. We analyzed public SARS-CoV-2 sequences and sequencing read data from ENA and identified the same deletion in over 100 genomes belonging to different lineages of SARS-CoV-2, pointing to a mutation hotspot or to positive selection. In almost all cases, the deletion was not represented in the virus genome sequence after consensus building. Additionally, further database searches point to other deletions in the ORF8 coding region that have never been reported by the standard data analysis pipelines. These findings and the fact that ORF8 is especially prone to deletions, make a clear case for the urgent necessity of public availability of the raw data for this and other large deletions that might change the physiology of the virus towards endemism.

Three novel corynebacterial species were isolated from soil sampled at a paddock in Vilsendorf, North Rhine-Westphalia, Germany. The strains were coccoid or irregular rod-shaped, catalase-positive and pale white to yellow-orange in colour. By whole genome sequencing and comparison of the 16S rRNA genes as well as the whole genome structure, it was shown that all three strains represent novel species of the family Corynebacteriaceae , order Corynebacteriales , class Actinobacteria . This project describes the isolation, identification, sequencing, and phenotypic characterization of the three novel Corynebacterium species. We propose the names Corynebacterium kalinowskii sp. nov. (DSM 110639T=LMG 31801T), Corynebacterium comes sp. nov. (DSM 110640T=LMG 31802T), and Corynebacterium occultum sp. nov. (DSM 110642T=LMG 31803T).

Motivation: Regions in intrinsic disordered proteins (IDPs) constitute important continuous aspects of protein function. While their existence on a structural continuum is widely accepted, most computational predictions have, nevertheless, focused on binary classifications. Existing datasets are severely limited in size and experimental evidence for continuous disorder. Results: Building on recently released datasets of continuous protein disorder and flexibility, we introduce UdonPred, a lightweight neural network exclusively inputting embeddings from the protein Language Model (pLM) ProstT5 to predict per-residue protein disorder from sequence alone. Training and evaluating UdonPred on seven datasets with divergent definitions of disorder and flexibility suggests that not model capacity, but agreement and nuance of disorder annotations, remains the main driver of performance. Binary disorder annotations can be reliably predicted from a multitude of different disorder and flexibility datasets, but there is still room for improvement in predicting continuous disorder. Availability: All code and data used for training and evaluation is available under an open-source license at https://github.com/davidwagemann/udonpred . Contact: assistant@rostlab.org Supplementary information: Supplementary data are available at Journal Name online.