Biology subjects

Schoenhuth, A.

Publications and source records attributed to Schoenhuth, A..

12 recordsLinked to original sources

bioRxiv2026

ANIMA: predicting protein-protein interactions across species

Motivation: Protein--protein interactions (PPIs) underpin a wide range of biological functions in living organisms. Experimental identification of new PPIs is expensive and time-consuming. The experimental bottleneck has implied an imbalance in terms of data availability: while certain species have been screened exhaustively, other species have not been sufficiently examined. An AI driven protocol for PPI prediction that leverages the massive data accumulated for certain species means a decisive boost for so far understudied species. Results: We present ANIMA (Artificial Neural Interaction Model for Animals), an AI supported cross-species PPI prediction model trained on popular species to predict PPIs in under-researched species. Our experiments demonstrate that our model, when trained on 200 diversely selected animal species, can successfully predict PPIs in other species: ANIMA achieves 95.3% accuracy on other animal species, 91.1% on other eukaryotes, and 82.5% on non-eukaryotes. For a more fine-grained evaluation of the model, we stratify performance rates by the evolutionary distance of test to training sets. We also stratify results by a novel, alignment based score ("representation score") which allows for fine-grained evaluation in terms of its capacity to generalize to unseen interactions. As expected, results demonstrate increasing performance on increasing evolutionary similarity and on increasing identity of interacting proteins, while still showing excellent performance on proteins entirely lacking counterparts in the training set. In comparison with the state of the art, ANIMA demonstrates substantial superiority in terms of performance rates.

bioinformatics↗

bioRxiv2026

Efficient and accurate near telomere-to-telomere haplotype reconstruction of diploid genomes

Telomere-to-telomere (T2T) and haplotype-resolved assembly are crucial for understanding eukaryotic genomes. For diploid species, this resolution is critical to uncover allelic variations, inheritance patterns, and functional genomic traits. Current scaffolding methods typically employ either sequence-based or graph-based strategies. Sequence-based approaches rely on proximity signals to yield high contiguity, but underutilize assembly graph information, resulting in more structural errors and chromosomal misassignments. Graph-based methods leverage graph topology for higher accuracy but frequently struggle to achieve chromosome-scale contiguity. However, neither strategy alone can overcome its inherent limitations to simultaneously achieve high contiguity and accuracy. To address these challenges, we introduce HapFold, the first hybrid scaffolding framework that synergistically leverages the complementary strengths of both graph-based and sequence-based approaches. By integrating the topological accuracy of assembly graphs with the proximity-guided contiguity of sequence models, HapFold achieves highly accurate, chromosome-scale or near-T2T haplotype reconstructions for diploid genomes. Compared to existing methods, HapFold achieves superior assembly quality while accelerating computation by an order of magnitude. Furthermore, in the haplotype reconstruction of diploid genomes using standard Oxford Nanopore Technologies simplex reads, HapFold enables the reconstruction of a greater number of near-T2T assemblies. Our approach provides a robust and scalable solution for the high-fidelity reconstruction of haplotype-resolved diploid genomes.

bioinformatics↗

bioRxiv2024

PhenoEncoder: A Discriminative Embedding Approach to Genomic Data Compression

AbstractExploring the heritability of complex genetic traits requires methods that can handle the genomes vast scale and the intricate re-lationships among genetic markers. Widely accepted association studies overlook non-linear effects (epistasis), prompting the adoption of deep neural networks (DNNs) for their scalability with large genetic datasets and ability to detect complex relationships. However, the curse of di-mensionality continues to limit the potential of DNNs, underscoring the critical need for dimensionality reduction for suitably sizing and shaping the genetic inputs, while preserving epistasis. Linkage disequilibrium (LD), a measure of correlation between genetic loci, offers a pathway to genome compression with minimal information loss. Using LD, the genome can be divided into smaller genomic regions, i.e., haplotype blocks, which can be locally compressed using deep au-toencoders. While autoencoders excel at preserving the main non-linear patterns, they still risk losing phenotype-relevant information when dom-inated by other sources of genetic variation. We propose a novel approach, PhenoEncoder, that incorporates pheno-typic variance directly into compression. This single nucleotide polymor-phism (SNP)-based pipeline employs multiple autoencoders, each dedi-cated to compressing a single haplotype block. The window-based spar-sity of the model eases the computational burden of simultaneously pro-cessing numerous SNPs. Concurrently, an auxiliary classifier predicts the phenotype from the compressed haplotype blocks. Epistasis is processed both within and between haplotype blocks by maintaining non-linearity in the autoencoders and the classifier. Through joint optimization of the compression and classification losses, PhenoEncoder ensures that disease-causing patterns are highlighted during compression. Applied to protein expression and simulated complex phenotype datasets, PhenoEncoder demonstrated enhanced generalizability in downstream classification tasks compared to standard autoencoder compression. By enabling phenotype-aware compression, PhenoEncoder emerges as a promis-ing approach for discriminative genomic feature extraction.

genomics↗

bioRxiv2023

Computing Linkage Disequilibrium Aware Genome Embeddings using Autoencoders

MotivationThe completion of the human genome has paved the way for genome-wide association studies (GWAS), which have already succeeded in explaining certain proportions of heritability. GWAS are not optimally suited to detect potential non-linear effects in disease risk, possibly hidden in non-additive interactions (epistasis). Alternative methods for epistasis detection using e.g. deep neural networks are currently under active development. Despite their promise to scale to high-dimensional data, deep neural networks (DNNs) are constrained by finite computational resources, which can be rapidly depleted due to increasing complexity with the sheer size of the genome. Besides, the curse of dimensionality complicates the task of capturing meaningful genetic patterns for DNNs; therefore calls for precaution in the form of dimensionality reduction. ResultsWe propose a method to compress genotyping data, involving single nucleotide polymorphisms (SNPs), while leveraging the LD structure in the genome and preserving non-linear relations among variants. This method involves clustering correlated SNPs into haplotype blocks and training per-block autoencoders that are able to learn a compressed representation of the blocks relevant genetic content. We provide an adjustable autoencoder design to accommodate diverse blocks and to bypass extensive hyperparameter tuning. We applied this method to genotyping data from Project MinE which involves a total of 23,209 ALS cases and 90,249 healthy controls. We compressed the haplotype blocks of an entire chromosome using our autoencoder-based approach, and show that this leads to more than 99% average test reconstruction accuracy - i.e. minimal information loss - while compressing the input to nearly 10% of the original size. We demonstrate that haplotype-block based autoencoders outperform their linear alternative Principal Component Analysis (PCA) by approximately 3% chromosome-wide accuracy of reconstructed variants. To the extent of our knowledge, our approach is the first to simultaneously leverage genome haplotype structure and deep neural networks for dimensionality reduction of genetic data. Availability and ImplementationData used in this study are available for academic use through the Project MinE Consortium at https://www.projectmine.com/research/data-sharing/. Data accessibility is contingent upon any terms or requirements specified by the source studies. Codes for data compression are available at https://github.com/gizem-tas/haploblock-autoencoders.

genomics↗

bioRxiv2021

phasebook: haplotype-aware de novo assembly of diploid genomes from long reads

Haplotype-aware diploid genome assembly is crucial in genomics, precision medicine, and many other disciplines. Long-read sequencing technologies have greatly improved genome assembly thanks to advantages of read length. However, current long-read assemblers usually introduce disturbing biases or fail to capture the haplotype diversity of the diploid genome. Here, we present phasebook, a novel approach for reconstructing the haplotypes of diploid genomes from long reads de novo. Benchmarking experiments demonstrate that our method outperforms other approaches in terms of haplotype coverage by large margins, while preserving competitive performance or even achieving advantages in terms of all other aspects relevant for genome assembly.

bioinformatics↗

bioRxiv2021

Strainline: full-length de novo viral haplotype reconstruction from noisy long reads

Haplotype-resolved de novo assembly of highly diverse virus genomes is critical in prevention, control and treatment of viral diseases. Current methods either can handle only relatively accurate short read data, or collapse haplotype-specific variations into consensus sequence. Here, we present Strainline, a novel approach to assemble viral haplotypes from noisy long reads without a reference genome. As a crucial consequence, Strainline is the first approach to provide strain-resolved, full-length de novo assemblies of viral quasispecies from noisy third-generation sequencing data. Benchmarking experiments on both simulated and real datasets of varying complexity and diversity confirm this novelty, by demonstrating the superiority of Strainline in terms of relevant criteria in comparison with the state of the art.

bioinformatics↗

bioRxiv2019

Viral quasispecies reconstruction via contig abundance estimation in variation graphs

The goal of strain-aware genome assembly is to reconstruct all individual haplotypes from a mixed sample at the strain level and to provide abundance estimates for the strains. Given that the use of a reference genome can introduce significant biases, de novo approaches are most suitable for this task. So far, reference-genome-independent assemblers have been shown to reconstruct haplotypes for mixed samples of limited complexity and genomes not exceeding 10000 bp in length. Here, we present VG-Flow, a de novo approach that enables full-length haplotype reconstruction from pre-assembled contigs of complex mixed samples. Our method increases contiguity of the input assembly and, at the same time, it performs haplotype abundance estimation. VG-Flow is the first approach to require polynomial, and not exponential runtime in terms of the underlying graphs. Since runtime increases only linearly in the length of the genomes in practice, it enables the reconstruction also of genomes that are longer by orders of magnitude, thereby establishing the first de novo solution to strain-aware full-length genome assembly applicable to bacterial sized genomes. VG-Flow is based on the flow variation graph as a novel concept that both captures all diversity present in the sample and enables to cast the central contig abundance estimation problem as a flow-like, polynomial time solvable optimization problem. As a consequence, we are in position to compute maximal-length haplotypes in terms of decomposing the resulting flow efficiently using a greedy algorithm, and obtain accurate frequency estimates for the reconstructed haplotypes through linear programming techniques. Benchmarking experiments show that our method outperforms state-of-the-art approaches on mixed samples from short genomes in terms of assembly accuracy as well as abundance estimation. Experiments on longer, bacterial sized genomes demonstrate that VG-Flow is the only current approach that can reconstruct full-length haplotypes from mixed samples at the strain level in human-affordable runtime.

bioinformatics↗

bioRxiv2019

Using the structure of genome data in the design of deep neural networks for predicting amyotrophic lateral sclerosis from genotype

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease caused by aberrations in the genome. While several disease-causing variants have been identified, a major part of heritability remains unexplained. ALS is believed to have a complex genetic basis where nonadditive combinations of variants constitute disease, which cannot be picked up using the linear models employed in classical genotype-phenotype association studies. Deep learning on the other hand is highly promising for identifying such complex relations. We therefore developed a deep-learning based approach for the classification of ALS patients versus healthy individuals from the Dutch cohort of the ProjectMinE dataset. Based on recent insight that regulatory regions on the genome play a major role in ALS, we employ a two-step approach: first promoter regions that are likely associated to ALS are identified, and second individuals are classified based on their genotype in the selected genomic regions. Both steps employ a deep convolutional neural network. The network architecture accounts for the structure of genome data by applying convolution only to parts of the data where this makes sense from a genomics perspective. Our approach identifies potential ALS-associated genetic variants, and generally outperforms other classification methods. Test results support the hypothesis that ALS is caused by non-additive combinations of variants. Our method can be applied to large-scale whole genome data. We consider this a first step towards genotype-phenotype association with deep learning that is tailored to genomics and can deal with genome-sized data.

genetics↗

bioRxiv2019

OGRE: Overlap Graph-based metagenomic Read clustEring

The microbes that live in an environment can be identified from the genomic material that is present, also referred to as the metagenome. Using Next Generation Sequencing techniques this genomic material can be obtained from the environment, resulting in a large set of sequencing reads. A proper assembly of these reads into contigs or even full genomes allows one to identify the microbial species and strains that live in the environment. Assembling a metagenome is a challenging task and can benefit from clustering the reads into species-specific bins prior to assembly. In this paper we propose OGRE, an Overlap-Graph based Read clustEring procedure for metagenomic read data. OGRE is the only method that can successfully cluster reads in species-specific bins for large metagenomic datasets without running into computation time-or memory issues.

bioinformatics↗

bioRxiv2018

Overlap graph-based generation of haplotigs for diploids and polyploids

Haplotype aware genome assembly plays an important role in genetics, medicine, and various other disciplines, yet generation of haplotype-resolved de novo assemblies remains a major challenge. Beyond distinguishing between errors and true sequential variants, one needs to assign the true variants to the different genome copies. Recent work has pointed out that the enormous quantities of traditional NGS read data have been greatly underexploited in terms of haplotig computation so far, which reflects that methodology for reference independent haplotig computation has not yet reached maturity. We present POLYTE (POLYploid genome fitTEr) as a new approach to de novo generation of haplotigs for diploid and polyploid genomes. Our method follows an iterative scheme where in each iteration reads or contigs are joined, based on their interplay in terms of an underlying haplotype-aware overlap graph. Along the iterations, contigs grow while preserving their haplotype identity. Benchmarking experiments on both real and simulated data demonstrate that POLYTE establishes new standards in terms of error-free reconstruction of haplotype-specific sequence. As a consequence, POLYTE outperforms state-of-the-art approaches in various relevant aspects, where advantages become particularly distinct in polyploid settings. POLYTE is freely available as part of the HaploConduct package at https://github.com/HaploConduct/HaploConduct, implemented in Python and C++.

bioinformatics↗

bioRxiv2018

Full-length de novo viral quasispecies assembly through variation graph construction

MotivationViruses populate their hosts as a viral quasispecies: a collection of genetically related mutant strains. Viral quasispecies assembly refers to reconstructing the strain-specific haplotypes from read data, and predicting their relative abundances within the mix of strains, an important step for various treatment-related reasons. Reference-genome-independent (\"de novo\") approaches have yielded benefits over reference-guided approaches, because reference-induced biases can become overwhelming when dealing with divergent strains. While being very accurate, extant de novo methods only yield rather short contigs. It remains to reconstruct full-length haplotypes together with their abundances from such contigs.\n\nMethodWe first construct a variation graph, a recently popular, suitable structure for arranging and integrating several related genomes, from the short input contigs, without making use of a reference genome. To obtain paths through the variation graph that reflect the original haplotypes, we solve a minimization problem that yields a selection of maximal-length paths that is optimal in terms of being compatible with the read coverages computed for the nodes of the variation graph. We output the resulting selection of maximal length paths as the haplotypes, together with their abundances.\n\nResultsBenchmarking experiments on challenging simulated data sets show significant improvements in assembly contiguity compared to the input contigs, while preserving low error rates. As a consequence, our method outperforms all state-of-the-art viral quasispecies assemblers that aim at the construction of full-length haplotypes, in terms of various relevant assembly measures. Our tool, Virus-VG, is publicly available at https://bitbucket.org/jbaaijens/virus-vg.

bioinformatics↗

bioRxiv2016

WhatsHap: fast and accurate read-based phasing

Read-based phasing allows to reconstruct the haplotypes of a sample purely from sequencing reads. While phasing is an important step for answering questions about population genetics, compound heterozygosity, and to aid in clinical decision making, there has been a lack of accurate, usable and standards-based software.\n\nWhatsHap is a production-ready tool for highly accurate read-based phasing. It was designed from the beginning to leverage third-generation sequencing technologies, whose long reads can span many variants and are therefore ideal for phasing. WhatsHap works also well with second-generation data, is easy to use and will phase not only SNVs, but also indels and other variants. It is unique in its ability to combine read-based with pedigree-based phasing, allowing to further improve accuracy if multiple related samples are provided.

bioinformatics↗