The Integrated Science Lab invites you to join the conversation at a Lunch Pitch. Kelly Swarts will talk about predicting missing growth in trees, Morgan Jones will discuss paleoclimate research opportunities in Svalbard, and Chinmay Dwibedi wants more representative genome coverage of the human gut microbiome.
To encourage cross pollination of ideas between researchers from different disciplines, IceLab hosts interdisciplinary research lunches with the vision of allowing ideas to meet and mate. During the Lunch Pitch Season, the creative lunches take place at KBC (Glasburen) on a Wednesday.
Register to come to the pitch and reserve your lunch by Monday, 4 May at 10am.
IceLab Lunch Pitch registration will open two weeks before the event.
Note! The default lunch option is a vegetarian falafel sandwich. You can choose an alternative lunch in a separate form that will be emailed to you once you have registered.
Predicting missing growth from tree increment cores
Utilizing annual growth data from tree increment cores requires the correct assignment of year of growth to each observed ring so that it can be associated with environmental records such as weather station or satellite data. However, sometimes trees do not grow because of unfavorable growing conditions, disease, etc., resulting in mis-categorized years for the rest of the series. Typically, identification of missing growth data, called “cross-dating”, is done by leveraging shared environmental responses between neighboring trees. Since missing rings are rarely shared by all trees for the same year, it is possible to identify missing rings through inference.
Unfortunately, the state of the art is that this is done by “people who are good at it”, which is not really replicable and accuracy is nearly impossible to assess. Dani Muraja in my group has been working on a Fourier phase based approach to comparing series, which can identify series with likely missing data, but cannot identify the problematic year reliably, primarily due to noisy signal with variable power over time.
We’ve been working on methods to scale the approach to populations using temporal (video) CNNs, but it’s not clear what the appropriate architecture is.
​
SvalClime: Deep-time paleoclimate archive opportunities in and around Svalbard
Paleoclimate is important to study as natural analogues to current and future change are only available in the geological record. Svalbard is an exceptional locality for deep-time paleoclimate research because 1) there is a sedimentary record somewhere on the archipelago that covers >80% of the last 600 million years; 2) it has modern transport links and logistics that allow for accessible field campaigns at 78 °N; and 3) it provides a rare high-latitude perspective on environmental change as existing records are dominated by lower-latitude localities. However, rock faces are often affected by the harsh climate conditions, which prevents high resolution study of past climate variability. The ideal study material is drill core, as this provides a near-continuous record that has been protected by the effects of surface weathering.
A MagellanPlus workshop in 2022 set up the SvalClime collaboration, which seeks to apply for funding opportunities to drill key environmental events in the Phanerozoic. This includes transitions to Icehouse worlds in the Late Paleozoic (360–255 million years ago [Ma]) and Eocene-Oligocene (34 Ma), but also periods of extreme warming known as hyperthermals that interspersed the Mesozoic (252 – 66 Ma). These drilling initiatives offer exciting opportunities for other scientific fields, such as the taxonomic and metabolic diversity of microbial communities in the intraterrestrial deep biosphere.
Opening the collaboration further expands the scientific potential of SvalClime, while also increasing the chances of success when applying for International Continental scientific Drilling Program (ICDP) funding.
Reconstruction of Alpha genomes of human gut bacteria
It is estimated that approximately 80% of human gut bacteria remain uncultured, leaving within-species diversity largely unexplored. Consequently, existing reference genomes do not adequately capture the full extent of global gene diversity. To address this limitation, we propose to construct gene-aware, graph-based reference genomes that better represent global microbial diversity and improve the mapping of gut metagenomic reads.
Interested in: Method development for graph-based assemblies combining sparse genes.
KBC Glasburen, near the KBC café. Find your way to the venue ()
IceLab Lunch Pitches are made possible through funding from KBC for the venue and from Stress Response Modeling at IceLab for their coordination and lunches.
Kelly Swarts, Assistant professor, UPSC
