Andreas Wilm

Summary: The exogenous, i.e. in vitro, loading of peptides onto major histocompatibility complex (MHC) class I molecules is a key step in many immunology-related experimental workflows. Here, we provide a machine learning solution, PIPLOM, which is specifically tailored to predict whether peptides can be loaded exogenously onto an MHC class I molecule. Benchmarking on 38 unseen epitopes with in-house ELISA (enzyme-linked immunosorbent assay) experiments showed that PIPLOM is outperforming well-established methods such as NETMHCpan-4.0 or MHCflurry, which are commonly used for the related task of predicting epitope HLA (human leukocyte antigen) haplotype specificity. Availability and implementation: Source code and data are available as Zenodo package 10.5281/zenodo.13771214. PIPLOM is available as a web service at https://piplom.immunoscape.com/.

The standardization, portability, and reproducibility of analysis pipelines is a renowned problem within the bioinformatics community. Most pipelines are designed for execution on-premise, and the associated software dependencies are tightly coupled with the local compute environment. This leads to poor pipeline portability and reproducibility of the ensuing results - both of which are fundamental requirements for the validation of scientific findings. Here, we introduce nf-core : a framework that provides a community-driven, peer-reviewed platform for the development of best practice analysis pipelines written in Nextflow. Key obstacles in pipeline development such as portability, reproducibility, scalability and unified parallelism are inherently addressed by all nf-core pipelines. We are also continually developing a suite of tools that assist in the creation and development of both new and existing pipelines. Our primary goal is to provide a platform for high-quality, reproducible bioinformatics pipelines that can be utilized across various institutions and research facilities.

Viral populations are complex, dynamic, and fast evolving. The evolution of groups of closely related viruses in a competitive environment is termed quasispecies. To fully understand the role that quasispecies play in viral evolution, characterizing the trajectories of viral genotypes in an evolving population is the key. In particular, long-range haplotype information for thousands of individual viruses is critical; yet generating this information is non-trivial. Popular deep sequencing methods generate relatively short reads that do not preserve linkage information, while third generation sequencing methods have higher error rates that make detection of low frequency mutations a bioinformatics challenge. Here we applied BAsE-Seq, an Illumina-based single-virion sequencing technology, to eight samples from four chronic hepatitis B (CHB) patients – once before antiviral treatment and once after viral rebound due to resistance. With single-virion sequencing, we obtained 248–8796 single-virion sequences per sample, which allowed us to find evidence for both hard and soft selective sweeps. We were able to reconstruct population demographic history that was independently verified by clinically collected data. We further verified four of the samples independently through PacBio SMRT and Illumina Pooled deep sequencing. Overall, we showed that single-virion sequencing yields insight into viral evolution and population dynamics in an efficient and high throughput manner. We believe that single-virion sequencing is widely applicable to the study of viral evolution in the context of drug resistance and host adaptation, allows differentiation between soft or hard selective sweeps, and may be useful in the reconstruction of intra-host viral population demographic history.

Viral populations are complex, dynamic, and fast evolving. The evolution of groups of closely related viruses in a competitive environment is termed quasispecies. To fully understand the role that quasispecies play in viral evolution, characterizing the trajectories of viral genotypes in an evolving population is the key. In particular, long-range haplotype information for thousands of individual viruses is critical; yet generating this information is non-trivial. Popular deep sequencing methods generate relatively short reads that do not preserve linkage information, while third generation sequencing methods have higher error rates that make detection of low frequency mutations a bioinformatics challenge. Here we applied BAsE-Seq, an Illumina-based single-virion sequencing technology, to eight samples from four chronic hepatitis B (CHB) patients – once before antiviral treatment and once after viral rebound due to resistance. We obtained 248-8,796 single-virion sequences per sample, which allowed us to find evidence for both hard and soft selective sweeps. We were also able to reconstruct population demographic history that was independently verified by clinically collected data. We further verified four of the samples independently on PacBio and Illumina sequencers. Overall, we showed that single-virion sequencing yields insight into viral evolution and population dynamics in an efficient and high throughput manner. We believe that single-virion sequencing is widely applicable to the study of viral evolution in the context of drug resistance, differentiating between soft or hard selective sweeps, and the reconstruction of intra-host viral population demographic history.

Background: Nanopore sequencing provides a rapid, cheap and portable real-time sequencing platform with the potential to revolutionize genomics. However, several applications are limited by relatively high single-read error rates (>10 %), including RNA-seq, haplotype sequencing and 16S sequencing. Results: We developed the Intramolecular-ligated Nanopore Consensus Sequencing (INC-Seq) as a strategy for obtaining long and accurate nanopore reads, starting with low input DNA. Applying INC-Seq for 16S rRNA-based bacterial profiling generated full-length amplicon sequences with a median accuracy >97 %. Conclusions: INC-Seq reads enabled accurate species-level classification, identification of species at 0.1 % abundance and robust quantification of relative abundances, providing a cheap and effective approach for pathogen detection and microbiome profiling on the MinION system.

Whole metagenome analysis has the potential to reveal functional triggers of skin diseases, but issues of cost, robustness and sampling efficacy have limited its application. Here, we have established an alternative, clinically practical and robust metagenomic analysis protocol and applied it to 80 skin microbiome samples epidemiologically stratified for atopic dermatitis (AD). We have identified distinct non-flare, baseline skin microbiome signatures enriched for Streptococcus and Gemella but depleted for Dermacoccus in AD-prone versus normal healthy skin. Bacterial challenge assays using keratinocytes and monocyte-derived dendritic cells established distinct IL-1-mediated, innate and Th1-mediated adaptive immune responses with Staphylococcus aureus and Staphylococcus epidermidis. Bacterial differences were complemented by perturbations in the eukaryotic community and functional shifts in the microbiome-wide gene repertoire, which could exacerbate a dry and alkaline phenotype primed for pathogen growth and inflammation in AD-susceptible skin. These findings provide insights into how the skin microbial community, skin surface microenvironment and immune system cross-modulate each other, escalating the destructive feedback cycle between them that leads to AD flare.

Realizing the democratic promise of nanopore sequencing requires the development of new bioinformatics approaches to deal with its specific error characteristics. Here we present GraphMap, a mapping algorithm designed to analyse nanopore sequencing reads, which progressively refines candidate alignments to robustly handle potentially high-error rates and a fast graph traversal to align long reads with speed and high precision (>95%). Evaluation on MinION sequencing data sets against short- and long-read mappers indicates that GraphMap increases mapping sensitivity by 10–80% and maps >95% of bases. GraphMap alignments enabled single-nucleotide variant calling on the human genome with increased sensitivity (15%) over the next best mapper, precise detection of structural variants from length 100 bp to 4 kbp, and species and strain-specific identification of pathogens using MinION reads. GraphMap is available open source under the MIT license at https://github.com/isovic/graphmap.