Biology subjects

Soeding, J.

Publications and source records attributed to Soeding, J..

27 records · Page 2Linked to original sources

bioRxiv2020

Deconvolving nuclesome binding energy from experimental errors

1Eukaryotic genomes are compacted into nucleosomes, 147-bp of DNA wrapped around histone proteins. Nucleosomes can hinder transcription factors and other DNA-binding proteins from accessing the genome. This competition at promoters and enhancers regulates gene expression. Therefore, a quantitative understanding of gene regulation requires the quantitative prediction of nucleosome binding affinities. However, little is known for certain about the sequence preference of nucleosomes. Here we develop an integrated model of nucleosome binding and genome-wide measurements thereof. Our model learns similar nucleosome sequence preferences from MNase-Seq and CC-Seq datasets. We find that modelling the positional uncertainty of MNase-Seq deconvolves the commonly described smooth 10-bp-periodic sequence preference into a position-specific pattern more closely resembling the pattern obtained from high-resolution CC-Seq data. By analysing the CC-Seq data we reveal the strong preference of A/T at +/- 3 bp from the dyad as an experimental bias. Our integrated model can separate this bias of CC-Seq from the true nucleosome binding preference. Our results show that nucleosomes have position-specific sequence preferences, which probably play an important role in their competition with transcription factors. Furthermore, our comparison of diverse datasets shows that the experimental biases have a similar strength as the signal of nucleosome-positioning measurements. Validating nucleosome models on experiments with similar biases overestimates their prediction quality of the true nucleosome binding. There are still many open questions about the sequence preference of nucleosomes and our approach will need to be extended to answer them. Only integrated models that combine the thermodynamics of nucleosome binding with experimental errors can deconvolve the two and learn the true preferences of nucleosomes.

molecular biology↗

bioRxiv2019

A high-throughput screen for transcription activation domains reveals their sequence characteristics and permits reliable prediction by deep learning

Transcription activation domains (ADs) are encoded by a wide range of seemingly unrelated amino acid sequences, making it difficult to recognize features that permit their dynamic behavior, fuzzy interactions and target specificity. We screened a large set of random 30-mer peptides for AD function and trained a deep neural network (ADpred) on the AD-positive and negative sequences. ADpred correctly identifies known ADs within protein sequences and accurately predicts the consequences of mutations. We show that functional ADs are (1) located within intrinsically disordered regions with biased amino acid composition, (2) contain clusters of hydrophobic residues near acidic side chains, (3) are enriched or depleted for particular dipeptide sequences, and (4) have higher helical propensity than surrounding regions. Taken together, our findings fit the model of "fuzzy" binding through hydrophobic protein-protein interfaces, where activator-coactivator binding takes place in a dynamic hydrophobic environment rather than through combinations of sequence-specific interactions.

molecular biology↗

bioRxiv2019

MetaEuk - sensitive, high-throughput gene discovery and annotation for large-scale eukaryotic metagenomics

BackgroundMetagenomics is revolutionizing the study of microorganisms and their involvement in biological, biomedical, and geochemical processes, allowing us to investigate by direct sequencing a tremendous diversity of organisms without the need for prior cultivation. Unicellular eukaryotes play essential roles in most microbial communities as chief predators, decomposers, phototrophs, bacterial hosts, symbionts and parasites to plants and animals. Investigating their roles is therefore of great interest to ecology, biotechnology, human health, and evolution. However, the generally lower sequencing coverage, their more complex gene and genome architectures, and a lack of eukaryote-specific experimental and computational procedures have kept them on the sidelines of metagenomics. ResultsMetaEuk is a toolkit for high-throughput, reference-based discovery and annotation of protein-coding genes in eukaryotic metagenomic contigs. It performs fast searches with 6-frame-translated fragments covering all possible exons and optimally combines matches into multi-exon proteins. We used a benchmark of seven diverse, annotated genomes to show that MetaEuk is highly sensitive even under conditions of low sequence similarity to the reference database. To demonstrate MetaEuks power to discover novel eukaryotic proteins in large-scale metagenomic data, we assembled contigs from 912 samples of the Tara Oceans project. MetaEuk predicted >12,000,000 protein-coding genes in eight days on ten 16-core servers. Most of the discovered proteins are highly diverged from known proteins and originate from very sparsely sampled eukaryotic supergroups. ConclusionThe open-source (GPLv3) MetaEuk software (https://github.com/soedinglab/metaeuk) enables large-scale eukaryotic metagenomics through reference-based, sensitive taxonomic and functional annotation.

bioinformatics↗

bioRxiv2019

Mechanisms of active regulation of biomolecular condensates

Liquid-liquid phase separation is a key organizational principle in eukaryotic cells, on par with intracellular membranes. It allows cells to concentrate specific proteins into condensates, increasing reaction rates and achieving switch-like regulation. However, it is unclear how cells trigger condensate formation or dissolution and regulate their sizes. We predict from first principles two mechanisms of active regulation by post-translational modifications such as phosphorylation: In enrichment-inhibition, the regulating modifying enzyme enriches in condensates and the modifications of proteins inhibit their interactions. Stress granules, Cajal bodies, P granules, splicing speckles, and synapsin condensates obey this model. In localization-induction, condensates form around an immobilized modifying enzyme, whose modifications strengthen protein interactions. Spatially targeted condensates formed during transmembrane signaling, microtubule assembly, and actin polymerization conform to this model. The two models make testable predictions that can guide studies into the many emerging roles of biomolecular condensates.

cell biology↗

bioRxiv2019

HH-suite3 for fast remote homology detection and deep protein annotation

BackgroundHH-suite is a widely used open source software suite for sensitive sequence similarity searches and protein fold recognition. It is based on pairwise alignment of profile Hidden Markov models (HMMs), which represent multiple sequence alignments of homologous sequences. ResultsWe developed a single-instruction multiple-data (SIMD) vectorized implementation of the Viterbi algorithm for profile HMM alignment and introduced various other speed-ups. This accelerated HHsearch by a factor 4 and HHblits by a factor 2 over the previous version 2.0.16. HHblits3 is ~10x faster than PSI-BLAST and ~20x faster than HMMER3. Jobs to perform HHsearch and HHblits searches with many query profile HMMs can be parallelized over cores and over servers in a cluster using OpenMP and message passing interface (MPI). The free, open-source, GNU GPL(v3)-licensed software is available at https://github.com/soedinglab/hh-suite. ConclusionThe added functionalities and increased speed of HHsearch and HHblits should facilitate their use in large-scale protein structure and function prediction, e.g. in metagenomics and genomics projects.

bioinformatics↗

bioRxiv2018

Synthetic protein alignments by CCMgen quantify noise in residue-residue contact prediction

Compensatory mutations between protein residues that are in physical contact with each other can manifest themselves as statistical couplings between the corresponding columns in a multiple sequence alignment (MSA) of the protein family. Conversely, high coupling coefficients predict residues contacts. Methods for de-novo protein structure prediction based on this approach are becoming increasingly reliable. Their main limitation is the strong systematic and statistical noise in the estimation of coupling coefficients, which has so far limited their application to very large protein families. While most research has focused on boosting contact prediction quality by adding external information, little progress has been made to improve the statistical procedure at the core. In that regard, our lack of understanding of the sources of noise poses a major obstacle. We have developed CCMgen, the first method for simulating protein evolution by providing full control over the generation of realistic synthetic MSAs with pairwise statistical couplings between residue positions. This procedure requires an exact statistical model that reliably reproduces observed alignment statistics. With CCMpredPy we also provide an implementation of persistent contrastive divergence (PCD), a precise inference technique that enables us to learn the required high-quality statistical models. We demonstrate how CCMgen can facilitate the development and testing of contact prediction methods by analysing the systematic noise contributions from phylogeny and entropy. For that purpose we propose a simple entropy correction (EC) strategy which disentangles the correction for both sources of noise. We find that entropy contributes typically roughly twice as much noise as phylogeny.

bioinformatics↗

bioRxiv2018

Reconstructing complex lineage trees from scRNA-seq data using MERLoT

Advances in single-cell transcriptomics techniques are revolutionizing studies of cellular differentiation and heterogeneity. Consequently, it becomes possible to track the trajectory of thousands of genes across the cellular lineage trees that represent the temporal emergence of cell types during dynamic processes. However, reconstruction of cellular lineage trees with more than a few cell fates has proved challenging. We present MERLoT (https://github.com/soedinglab/merlot), a flexible and user-friendly tool to reconstruct complex lineage trees from single-cell transcriptomics data and further impute temporal gene expression profiles along the reconstructed tree structures. We demonstrate MERLoTs capabilities on various real cases and hundreds of simulated datasets.

bioinformatics↗

bioRxiv2018

PROSSTT: probabilistic simulation of single-cell RNA-seq data for complex differentiation processes

BackgroundSingle-cell RNA sequencing (scRNA-seq) is an enabling technology for the study of cellular differentiation and heterogeneity. From snapshots of the transcriptomic profiles of differentiating single cells, the cellular lineage tree that leads from a progenitor population to multiple types of differentiated cells can be derived. The underlying lineage trees of most published datasets are linear or have a single branchpoint, but many studies with more complex lineage trees will soon become available. To test and further develop tools for lineage tree reconstruction, we need test datasets with known trees.\n\nResultsPROSSTT can simulate scRNA-seq datasets for differentiation processes with lineage trees of any desired complexity, noise level, noise model, and size. PROSSTT also provides scripts to quantify the quality of predicted lineage trees.\n\nAvailabilityhttps://github.com/soedinglab/prosstt\n\nContactsoeding@mpibpc.mpg.de

bioinformatics↗

bioRxiv2017

Bayesian multiple logistic regression for meta-analyses of GWAS

Genetic variants in genome-wide association studies (GWAS) are tested for disease association mostly using simple regression, one variant at a time. Standard approaches to improve power in detecting disease-associated SNPs use multiple regression with Bayesian variable selection in which a sparsity-enforcing prior on effect sizes is used to avoid overtraining and all effect sizes are integrated out for posterior inference. For binary traits, the logistic model has not yielded clear improvements over the linear model. For multi-SNP analysis, the logistic model required costly and technically challenging MCMC sampling to perform the integration.\n\nHere, we introduce the quasi-Laplace approximation to solve the integral and avoid MCMC sampling. We expect the logistic model to perform much better than multiple linear regression except when predicted disease risks are spread closely around 0.5, because only close to its inflection point can the logistic function be well approximated by a linear function. Indeed, in extensive benchmarks with simulated phenotypes and real genotypes, our Bayesian multiple LOgistic REgression method (B-LORE) showed considerable improvements (1) when regressing on many variants in multiple loci at heritabilities [≥] 0.4 and (2) for unbalanced case-control ratios.\n\nB-LORE also enables meta-analysis by approximating the likelihood functions of individual studies by multivariate normal distributions, using their means and covariance matrices as summary statistics. Our work should make sparse multiple logistic regression attractive also for other applications with binary target variables. B-LORE is freely available from: https://github.com/soedinglab/b-lore.

genetics↗