Biology subjects

Haak, M.

Publications and source records attributed to Haak, M..

4 recordsLinked to original sources

bioRxiv2026

UdonPred: Untangling Protein Intrinsic Disorder Prediction

MotivationRegions 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. ResultsBuilding 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. AvailabilityAll code and data used for training and evaluation is available under an open-source license at https://github.com/davidwagemann/udonpred. Contactassistant@rostlab.org Supplementary informationSupplementary data are available at Journal Name online.

bioinformatics↗

bioRxiv2024

Discovery of a pentose as a cytosine nucleobase modification in Shewanella phage Thanatos-1 genomic DNA mediating enhanced resistance towards host restriction systems

Co-evolution of bacterial defense systems and phage counter defense mechanisms has resulted in an intricate biological interplay between bacteriophages and their prey. To evade nuclease-based mechanisms targeting the DNA, various bacteriophages modify their nucleobases, which impedes or even inhibits recognition by endonucleases. We found that Shewanella phage Thanatos-1 DNA is insensitive to multiple restriction enzymes and, partially, also to Cas I-Fv and Cas9 cleavage. Furthermore, the phage genome shows strongly impaired basecalling with nanopore sequencing. We characterised the phage adenine methyltransferase TH1_126 in methylase-free E. coli ER3413 and derived and confirmed its recognition motif 5-ATC-3. Moreover, the data pointed to an additional, much more substantial nucleobase modification. Using LC-MS, we identified a deoxypentose of unknown configuration attached to cytosine as a yet undiscovered phage DNA modification, which is present in Thanatos-1 genomic DNA, likely mediates the observed resistance to restriction endonucleases, as well as a strong reduction in Cas nuclease activity. To elucidate the underlying enzyme functions, we determined structural homologs of Thanatos-1 proteins among known glycosyltransferase folds and experimentally proved a UDP-xylose pyrophosphorylase function of phage protein TH1_063 by in vitro enzyme assays.

molecular biology↗

bioRxiv2023

Deletion of SMC renders FtsK essential in Corynebacterium glutamicum

Structural maintenance of chromosomes (SMC) are ubiquitously distributed proteins involved in chromosome organization. Deletion of smc causes severe growth phenotypes in many organisms. Surprisingly, smc can be deleted in Corynebacterium glutamicum, a member of the Actinomycetota phylum, without any apparent growth phenotype. Earlier work has shown that SMC in C. glutamicum is loaded in a ParB-dependent fashion to the chromosome and functions in replichore cohesion. The unexpected absence of a growth phenotype in the smc mutant prompted us to screen for unknown synthetic interactions within C. glutamicum. Therefore, we generated a high-density Tn-5 library based on wild-type and smc-deleted C. glutamicum strains. The transposon sequencing (Tn-seq) data revealed that the DNA-translocase FtsK is essential in a smc deletion strain. FtsK localized to the septa and cell poles in wild type cells, however deletion of smc resulted in a decreased polar FtsK localization. Single-particle tracking analysis further suggests that prolonged FtsK complex activity is both required and sufficient to make up for the absence of SMC, thus achieving efficient chromosome segregation in C. glutamicum. Further, single molecule dynamics of FtsK is influenced, albeit indirectly, by DNA-loaded SMC. Deletion of ParB results in an increased of both SMC and FtsK mobility. While the first change agrees with previous data that show how ParB is essential for SMC loading on DNA, the latter suggests that FtsK mobility is affected in cells with defects in chromosome organization. Based on our data we propose a simple, yet efficient mechanism for efficient DNA segregation in C. glutamicum, even in absence of SMC proteins. ImportanceFaithful DNA segregation is of fundamental importance for life. Bacteria have efficient systems to coordinate chromosome compaction, DNA segregation and cell division. A key factor in DNA compaction is the SMC-complex that is found to be essential in many bacteria. In members of the Actinomycetota smc is dispensable, but the reason for this was unclear. We show here that the divisome associated DNA-pump FtsK can compensate SMC loss and the subsequent loss in correct chromosome organization. In cells with distorted chromosomes, FtsK functions for an extended period of time at the septum, until chromosomes are segregated.

microbiology↗

bioRxiv2019

Nanopore sequencing of native adeno-associated virus single-stranded DNA using a transposase-based rapid protocol

Next-generation sequencing of single-stranded DNA (ssDNA) enables transgene characterization of gene-therapy vectors such as adeno-associated virus (AAV), but current library generation uses complicated and potentially biased second-strand synthesis. We report that libraries for nanopore sequencing of ssDNA can be conveniently created without second-strand synthesis using a transposase-based protocol. We show for bacteriophage M13 ssDNA that the MuA transposase has unexpected residual activity on ssDNA, explained in part by transposase action on transient double-stranded hairpins. In case of AAV, library creation is additionally aided by genome hybridization. We demonstrate the power of direct sequencing combined with nanopore long reads by characterizing AAV vector transgenes. Sequencing yielded reads up to full genome length, including GC-rich inverted terminal repeats. Unlike short-read techniques, single reads covered genome-genome and genome-contaminant fusions and other recombination events, while additionally providing information on epigenetic methylation. Single-nucleotide variants across the transgene cassette were revealed and secondary genome packaging signals were readily identified. Moreover, comparison of sequence abundance with qPCR results demonstrated the techniques future potential for quantification of DNA impurities in AAV vector stocks. The findings promote direct nanopore sequencing as a fast and versatile platform for ssDNA characterization, such as AAV ssDNA in research and clinical settings.

genomics↗