Michael Sars Symposium 2026

This schedule is tentative and will be updated regularly. Abstract for the talks are further down on this page. 

List of Abstracts

The molecular machines that transport eukaryotic lipids
Anne-Claude Gavin, Université de Genève

Eukaryotic cells produce thousands of different lipids—collectively known as the lipidome—whose composition is tailored to cellular needs. Lipids are distributed unevenly throughout biological systems, where they accumulate locally, forming membranes with specific compositions and thereby determining the identity and functional specialization of organelles. Due to their hydrophobicity, lipids cannot move freely out of cellular membranes through the cell’s aqueous environment and require transporters—lipid transfer proteins, or LTPs—to carry them. LTPs are soluble molecular machines responsible for transporting lipids, and they are found in all kingdoms of life. They have diverse structures, but many share a common mode of action: they extract specific lipids from membrane bilayers and load them into a hydrophobic pocket, forming water-soluble protein-lipid complexes that isolate cargoes from the aqueous phase. In addition to their cargo, some LTPs mobilize auxiliary lipids that function as exchange currencies or cofactors. They facilitate the uptake or release of the cargo, which would determine the direction of transport and its coupling to metabolism. However, for most LTPs, the identity of cargo and auxiliary lipids remains unknown. The fundamental biochemistry of LTPs remains poorly understood, limiting our ability to study how they function within cells. Our goal is to begin addressing this gap; I will present a recent systematic analysis of the lipid binding properties of human LTPs and discuss the general principles that we have derived from it.
 

The Modern Mollusc: Integrating Embryology and Genomics to Understand an Animal Phylum
Deirdre Lyons, Scripps Institution of Oceanography

Molluscs are familiar invertebrates, ranging from the humble garden slug and the vibrant seashells adorning beaches to the intelligent, shell-less octopus. As one of the largest, most diverse, and visually striking groups of marine animals, molluscs have been used by humans for centuries, valued for materials like pearls and as culinary staples such as pulpo, escargot, and moules. Marine molluscs have also been pivotal in biological research. Notably, studies on a squid species led to the 1963 Nobel Prize in Physiology or Medicine for elucidating the ionic mechanisms of action potentials, and work with a sea slug species contributed to the 2000 Nobel Prize for discovering the biological mechanisms of memory storage. Despite these contributions, biomedicine has since increasingly shifted its focus to cellular and genetic levels, and no mollusc species has joined the pantheon of established "model" organisms like the fly, roundworm, zebrafish, or mouse. This is largely because commonly studied mollusc species were typically large, challenging to breed in laboratory settings, and, critically, their embryos were not amenable to genetic transformation—meaning exogenous DNA and RNA molecules could not be effectively delivered to their eggs, a prerequisite for techniques like CRISPR-Cas9 genome editing. In this talk, I will describe both the classical embryological studies that revealed how molluscs’ body plans are constructed, as well as the recent advancements to establish genetically tractable species, including snails, nudibranchs, and cephalopods. Using a comparative approach to understand the developmental processes that influence molluscan diversity, I will share recent insights into the evolutionary origins of conserved and novel features. These studies are ushering in a modern era of molluscan research, providing new tools to investigate the fascinating biology of this phylum.
 

Transcriptomic and cellular basis of developmental thermal stress in the ascidian Ciona intestinalis
Jonas Brandenburg, Michael Sars Centre, University of Bergen

Multicellular organisms need to yield consistent developmental outcomes in the context of a variable environment. Therefore, patterns of gene expression are tightly regulated. Deviations from the optimal developmental temperature can be buffered in a certain permissive range, leading to a temporal scaling of developmental speed, but will become detrimental outside of this range. As the effect of thermal stress is often cell type-specific, investigations of the effects of such environmental perturbations need to be performed at a scale and resolution matching this non-uniform response, e.g. by using single-cell genomics.

We leveraged a norwegian population of the sea squirt Ciona intestinalis to investigate developmental thermal effects at the single-cell level. We constructed a thermal profile for this population and generated a de novo genome using long-read sequencing. Finally, we profiled over 80 000 cells of embryos developing at the reference temperature of 15˚C. This highly continuous atlas revealed the transcriptomic and cellular dynamics of Ciona development.

Using a second scRNA-seq dataset (about 85 000 cells) of Ciona embryos grown at various permissive and non-permissive temperatures (12 to 21˚C) and sampled at developmental times matching the continous developmental atlas previously created, we investigated temperature-sensitive developmental dynamics in Ciona. In this temperature experiment, only the highest temeprature (21°C) exhibited markedly reduced hatching rates and increased levels of developmental deformations, while the time from fertilization to hatching scaled from 36h (12°C) to 19h (21°C). While broad celltypes identities appear to be maintained even in strongly perturbed embryos, more subtle, celltype-specific effects were appearent. Using cell-type abundance, developmental progression and differential gene expression as quantitative metrics, we identify cell type-specific sensitivty to temperature stress with notochord cells appearing among the most perturbed cells. Finally, gene expression changes in response to temperature treatments are widespread and often cell type-specific and can further be releated to common phenotypic.

Future work will aim to combine this dataset with high-content morphometric data to relate the transcriptional effects of temperature to its morphometric outcomes and to identify the underlying causes of temperature sensitivity in gene expression.

The genetic architecture of larval evolution
Matthew Rockman, New York University

Marine larvae are subject to the vagaries of the planktonic realm, where the densities of food and predators very enormously. Divergent conditions favor alternative life histories: abundant food and low predation select for individuals that produce large numbers of small larvae that feed in the plankton until undergoing a metamorphosis, while scarce food and abundant predation select for individuals that produce small numbers of large larvae that can metamorphose out of the plankton without having to feed there. One species, the polychaete annelid Streblospio benedicti, has evolved to exhibit both developmental modes, with each female producing one type of offspring or the other as a result of genetic differences. We use quantitative and population genetic methods to show that this dimorphism depends on a large number of discrete loci spread across the genome, sprinkled across a background of low differentiation. The surprising genetic architecture implicates incredibly intense selection in maintaining the dimorphism, and helps pinpoint the genes responsible for evolutionary changes in oogenesis and larval development.

It sounds simple: The phenotype is the gene in its environment
Christian Jørgensen, Department of Biological Sciences, University of Bergen

Roughly half a century ago, biologists realized how the phenotype can be thought of as a survival machine: adaptive evolution led to genes that build a phenotype that successfully propagates them to future generations. With DNA now in the driver’s seat, the life sciences saw a tremendous acceleration driven by ever new molecular and computational tools, but two big challenges remain. First, the genotype-to-phenotype mapping is muddled by a highly complex intracellular environment and by many traits being polygenic. Second, the performance of the phenotype further depends on the external environment, including the myriad other species living there. I argue that one can make advancements towards both these challenges by focusing more on the phenotype and its ecology, including flexible responses to variable environments. Like a Necker cube that one can see equally well from two perspectives, science needs to integrate the top-down perspective of natural selection with the bottom-up view of matter and energy, with information closing the loop.
 

Towards high-resolution mapping of TF-DNA interactions and TF cooperation across species
Anthony Mathelier, Norwegian Centre for Molecular Biosciences and Medicine (NCMBM), University of Oslo

Unlocking how gene expression is regulated is critical to understanding development, evolution, and disease. Transcription factors (TFs) are key proteins involved in gene expression regulation as they modulate when, where, and at which intensity genes are activated or repressed. Specifically, TFs bind cis-regulatory regions at TF binding sites (TFBSs) to control transcription. Importantly, TFs do not act individually but rather cooperate to control gene expression, either forming dimers or interacting on the same cis-regulatory regions. Hence, understanding where TFs bind to the genome and cooperate is critical for decoding the cis-regulatory grammar encoded in genomes. In this talk, I will provide an overview of our recent efforts towards high-resolution mapping of TF-DNA interactions and TF cooperation across species, including the development of the JASPAR and UniBind databases, and how these resources help use understand transcription regulation in health and disease.
 

Long-range transcriptional regulation of multicellular processes: genomic organisation and evolutionary origins
Boris Lenhard, Imperial College London

Clusters of extremely conserved non-coding elements (CNEs) are a prominent feature of Metazoan genomes. They span the loci of genes whose products regulate multicellular development, including developmental transcription factors, signalling proteins, neural and developmental cell adhesion proteins, and axon guidance molecules. Many CNEs act as enhancers, and their organisation into genomic regulatory blocks (GRBs) suggests that these loci represent an ancient regulatory architecture for coordinating cell fate decisions across developing multicellular structures.

In the talk, I will present a set of criteria we developed to define the boundaries of regions containing regulatory elements of developmental genes and to identify the candidate target genes within them. This is essential because CNEs frequently lie inside or beyond unrelated bystander genes, while their regulatory activity is directed towards distant developmental target promoters. The functional identity of GRB target genes is highly constrained, even when the size of the surrounding locus differs by orders of magnitude across genomes. Equivalent regulatory domains may span 10-20 kb in nematodes, hundreds of kb in insects, 1-2 Mb in mammals, and more than 10 Mb in species with highly expanded genomes such as axolotl or lungfish.

I will also describe new approaches we developed for matching orthologous CNEs across large evolutionary distances. The results indicate that many CNEs are older than previously recognised, that their sequence-level conservation is progressively obscured by slow turnover after lineage separation, and that inclusion of lineages with short branches is key to more complete reconstruction of ancestral regulatory elements. This supports a model in which ancient clusters of enhancers were already present around multicellularity-coordinating genes in early animal ancestors, and have since been maintained as regulatory systems even when individual enhancer sequences were lost, replaced, or became unrecognisable.

Finally, I will discuss mounting evidence that related forms of conserved long-range regulatory organisation may also exist in multicellular plants and some fungal lineages. If correct, these findings would suggest that this regulatory mechanism, and perhaps the regulatory requirements of complex multicellularity itself, are older than is usually assumed.
 

Causal inference in systems genetics
Tom Michoel, Computational Biology Unit, University of Bergen

Complex traits and diseases arise from genetic and environmental perturbations that propagate through molecular networks across tissues. In this talk, I will discuss how causal inference in systems genetics uses genetic variation as a natural experiment to disentangle cause from correlation in multi‑omics data. I will introduce approaches based on Mendelian randomization and causal Bayesian networks to reconstruct gene and protein networks, identify disease‑driving pathways, and link molecular variation to phenotype. Using cardiovascular disease as a case study, I will show how circulating biomarkers can reveal cross‑tissue disease mechanisms and enable biologically interpretable risk prediction.
 

Exploring critical and bifurcation points across biological contexts: from tuning to function
Isabella Graf, EMBL Heidelberg and Heidelberg University

Many living systems demonstrate exquisite sensitivity to small input signals. A tempting hypothesis is that these systems operate close to bifurcation or critical points, where the system's response exhibits a diverging susceptibility to the control parameter and small signals are amplified into a large collective response. A common concern, however, is that proximity to such points requires fine-tuning of parameters, which seems impossible for noisy biological systems. Based on several distinct sensory systems, we have investigated a feedback motif that robustly maintains these systems close to their respective bifurcation point. The key ingredient is that the collective response feeds back onto the control parameter. To illustrate this idea, I will mention several examples ranging from snake thermosensing to mammalian hearing and discuss the functional benefits associated with being near-critical.
 

Markov state modelling of stochastic gene expression
Susanna Röblitz, Computational Biology Unit, University of Bergen

Gene expression is a stochastic process, leading to cell-to-cell variation in mRNA and protein levels even in genetically identical cells under the same conditions. One reason for this noise is the low number of molecules involved in the process, such as a single gene or mRNA transcript, which makes molecular events like binding and decay inherently random and unpredictable. This randomness often results in complex multi-attractor dynamics, where attractors can for example be identified with committed and primed states in cell differentiation. One way to describe the long-time behavior of such systems is to run stochastic simulations, but the occurrence of rare events typically demands very long and extensive simulations. In this presentation, I’ll outline an alternative approach known as Markov state modeling (MSM). MSM enables the efficient simulation of rare events and provides a description of cellular phenotypes in terms of their gene expression and transition patterns.