Biology subjects

Ecker, N.

Publications and source records attributed to Ecker, N..

5 recordsLinked to original sources

bioRxiv2025

The role of plant polyploidy in the structure of plant-pollinator communities

Polyploidization is a major macromutation, bearing notable genomic and ecological consequences. While the impact of polyploidy on plant abiotic niches is well studied, our understanding of its consequences on biotic interactions, and particularly pollination, is lacking and hardly considers its role in shaping community structure. Here, we integrate hundreds of plant-pollinator networks, ploidy inferences, reproductive traits, and climatic attributes to ascertain whether a general pattern characterizes the link between polyploid frequency and community structure. We further examine whether environmental factors and plant traits known to be associated with polyploidy mediate this relationship. Our analysis reveals that an increased frequency of polyploid species is positively associated with network nestedness while being negatively associated modularity. Structural equation modeling reveals that these associations are partially mediated via the frequency of self-compatible plants and to a lesser extent by differences in flower shape. Despite these alterations in community structure, the heightened abundance of polyploids appears to have minimal impact on network connectance and resilience to extinction. Our findings imply that unlike abiotic interactions, the impact of polyploidy on biotic interactions is less predictable, and is affected synergistically by both phenotypic and environmental factors. Collectively, however, polyploidy still exerts an influence on community structure.

evolutionary biology↗

bioRxiv2024

BetaAlign: a deep learning approach for multiple sequence alignment

The multiple sequence alignment (MSA) problem is a fundamental pillar in bioinformatics, comparative genomics, and phylogenetics. Here we characterize and improve BetaAlign, the first deep learning aligner, which substantially deviates from conventional algorithms of alignment computation. BetaAlign draws on natural language processing (NLP) techniques and trains transformers to map a set of unaligned biological sequences to an MSA. We show that our approach is highly accurate, comparable and sometimes better than state-of-the-art alignment tools. We characterize the performance of BetaAlign and the effect of various aspects on accuracy; for example, the size of the training data, the effect of different transformer architectures, and the effect of learning on a subspace of indel-model parameters (subspace learning). We also introduce a new technique that leads to improved performance compared to our previous approach. Our findings further uncover the potential of NLP-based approaches for sequence alignment, highlighting that AI-based methodologies can substantially challenge classic tasks in phylogenomics and bioinformatics.

bioinformatics↗

bioRxiv2022

Harnessing machine translation methods for sequence alignment

The sequence alignment problem is one of the most fundamental problems in bioinformatics and a plethora of methods were devised to tackle it. Here we introduce BetaAlign, a novel methodology for aligning sequences using a natural language processing (NLP) approach. BetaAlign accounts for the possible variability of the evolutionary process among different datasets by using an ensemble of transformers, each trained on millions of samples generated from a different evolutionary model. Our approach leads to outstanding alignment accuracy, often outperforming commonly used methods, such as MAFFT, DIALIGN, ClustalW, T-Coffee, and MUSCLE. Notably, the utilization of deep-learning techniques for the sequence alignment problem brings additional advantages, such as automatic feature extraction that can be leveraged for a variety of downstream analysis tasks.

evolutionary biology↗

bioRxiv2022

Mechanics of human embryo compaction

The shaping of the human embryo begins with compaction, during which cells come into close contact and form a tighter structure1,2. Assisted reproductive technology (ART) studies suggest that human embryos fail compaction primarily because of defective adhesion3,4. Based on our current understanding of animal morphogenesis5,6, other morphogenetic engines, such as cell contractility, could be involved in shaping the human embryo. However, the molecular, cellular and physical mechanisms driving human embryo morphogenesis remain uncharacterized. Using micropipette aspiration on human embryos donated to research, we have mapped cell surface tensions during compaction. This reveals a 4-fold increase of tension at the cellmedium interface while cell-cell contacts keep a steady tension. Comparison between human and mouse reveals qualitatively similar but quantitively different mechanical strategies, with human embryos being mechanically least efficient. Inhibition of cell contractility and cell-cell adhesion in human embryos reveal that only contractility controls the surface tension responsible for compaction. Interestingly, if both cellular processes are required for compaction, they exhibit distinct mechanical signatures when faulty. Analyzing the mechanical signature of naturally failing embryos, we find evidence that non-compacting embryos or partially compacting embryos with excluded cells have defective contractility. Together, our study reveals that an evolutionarily conserved increase in cell contractility is required to generate the forces driving the first morphogenetic movement shaping the human body.

developmental biology↗

bioRxiv2020

Harnessing machine learning to unravel protein degradation in Escherichia coli

Degradation of intracellular proteins in Gram-negative bacteria regulates various cellular processes and serves as a quality control mechanism by eliminating damaged proteins. To understand what causes the proteolytic machinery of the cell to degrade some proteins while sparing others, we employed a quantitative pulsed-SILAC (Stable Isotope Labeling with Amino acids in Cell culture) method followed by mass spectrometry analysis to determine the half-lives for the proteome of exponentially growing Escherichia coli, under standard conditions. We developed a likelihood-based statistical test to find actively degraded proteins, and identified dozens of novel proteins that are fast-degrading. Finally, we used structural, physicochemical and protein-protein interaction network descriptors to train a machine-learning classifier to discriminate fast-degrading proteins from the rest of the proteome. Our combined computational-experimental approach provides means for proteomic-based discovery of fast degrading proteins in bacteria and the elucidation of the factors determining protein half-lives and have implications for protein engineering. Moreover, as rapidly degraded proteins may play an important role in pathogenesis, our findings could identify new potential antibacterial drug targets.

microbiology↗