Repository logo

DalSpace Institutional Repository

DalSpace is a digital service that collects, preserves, and distributes digital material produced by the Dalhousie community.

  • To learn about content guidelines, policies, and how to deposit, view the Help documents.
  • Contact us to get started submitting content to DalSpace at dalspace@dal.ca

Contact Us | Send Feedback

 

Recent Submissions

ItemOpen Access
LRRTM4 is a member of the trans-synaptic complex between rod photoreceptors and ON bipolar cells
(2021) Agosto, MA; Wensel, TG
Leucine rich repeat transmembrane (LRRTM) proteins are synaptic adhesion molecules with roles in synapse formation and signaling. LRRTM4 transcripts were previously shown to be enriched in rod bipolar cells (BCs), secondary neurons of the retina that form synapses with rod photoreceptors. Using two different antibodies, LRRTM4 was found to reside primarily at rod BC dendritic tips, where it colocalized with the transduction channel protein, TRPM1. LRRTM4 was not detected at dendritic tips of ON-cone BCs. Following somatic knockout of LRRTM4 in BCs by subretinal injection and electroporation of CRISPR/Cas9, LRRTM4 was abolished or reduced in the dendritic tips of transfected cells. Knockout cells had a normal complement of TRPM1 at their dendritic tips, while GPR179 accumulation was partially reduced. In experiments with heterologously expressed protein, the extracellular domain of LRRTM4 was found to engage in heparan-sulfate dependent binding with pikachurin. These results implicate LRRTM4 in the GPR179-pikachurin-dystroglycan transsynaptic complex at rod synapses.
ItemOpen Access
The mGluR6 ligand-binding domain, but not the C-terminal domain, is required for synaptic localization in retinal ON-bipolar cells
(2021) Agosto, MA; Adeosun, AAR; Kumar, N; Wensel, TG
Signals from retinal photoreceptors are processed in two parallel channels-the ON channel responds to light increments, while the OFF channel responds to light decrements. The ON pathway is mediated by ON type bipolar cells (BCs), which receive glutamatergic synaptic input from photoreceptors via a G-protein-coupled receptor signaling cascade. The metabotropic glutamate receptor mGluR6 is located at the dendritic tips of all ON-BCs and is required for synaptic transmission. Thus, it is critically important for delivery of information from photoreceptors into the ON pathway. In addition to detecting glutamate, mGluR6 participates in interactions with other postsynaptic proteins, as well as trans-synaptic interactions with presynaptic ELFN proteins. Mechanisms of mGluR6 synaptic targeting and functional interaction with other synaptic proteins are unknown. Here, we show that multiple regions in the mGluR6 ligand-binding domain are necessary for both synaptic localization in BCs and ELFN1 binding in vitro. However, these regions were not required for plasma membrane localization in heterologous cells, indicating that secretory trafficking and synaptic localization are controlled by different mechanisms. In contrast, the mGluR6 C-terminus was dispensable for synaptic localization. In mGluR6 null mice, localization of the postsynaptic channel protein TRPM1 was compromised. Introducing WT mGluR6 rescued TRPM1 localization, while a C-terminal deletion mutant had significantly reduced rescue ability. We propose a model in which trans-synaptic ELFN1 binding is necessary for mGluR6 postsynaptic localization, whereas the C-terminus has a role in mediating TRPM1 trafficking. These findings reveal different sequence determinants of the multifunctional roles of mGluR6 in ON-BCs.
ItemOpen Access
ADDRESSING CLASS IMBALANCE IN HIERARCHICAL MULTILABEL CLASSIFICATION WITH NODE-WISE WEIGHTING
(2025-06-13) Sharma, Ayushi; Not Applicable; Master of Computer Science; Faculty of Computer Science; Not Applicable; N/A; Not Applicable; Dr. Israat Haque; Dr. Gabriel Spadon De Souza; Dr. Thomas Trappenberg
Class imbalance is a persistent challenge in hierarchical multilabel (HML) classification tasks where labels form parent-child relationships and samples contain more than one label. It hampers learning as the majority classes dominate training while minority classes are neglected, which is amplified due to hierarchical dependencies. We introduce an adapted node-wise weighting method that enforces hierarchical constraints and reimagine HML imbalance as defined by node frequencies in a dataset. We benchmark it on several HML datasets including functional genomics datasets and an oceanographic dataset. We analyze which evaluation metrics provide a comprehensive assessment that is specifically suited for assessing HML tasks. The experimental results show that our node-wise weighting method consistently improves recall for minority classes without sacrificing much precision on majority classes, outperforming prior techniques for handling class imbalance. These findings show the potential of our method to address class imbalance in HML settings applicable to diverse real-world HML tasks.
ItemEmbargo
Assessment of a foundational machine-learned potential for energy ranking of molecular crystal polymorphs
(Royal Society of Chemistry, 2025-05-19) Nickerson, Cameron J.; Johnson, Erin R.
ItemOpen Access
Sustaining Soil Organic Carbon and Soil Health in Intensively Tilled Short Crop Rotations
(2025-05-29) Lavergne, Stéphanie; No; Doctor of Philosophy; Department of Plant, Food and Environmental Sciences; Received; Meagan E. Schipanski; Yes; Andrew M. Hammermeister; David L. Burton; Derek H. Lynch; Caroline Halde
Soils in Canada are under degradation due to the intensification of agricultural systems. Intensive field crop rotations are characterized by low crop diversity and crop residues, and high tillage frequency and external nutrient inputs. However, the impact of these intensive short rotations on soil health, earthworm communities, and soil organic carbon (SOC) has been little investigated and is the focus of this thesis. A field survey was conducted over three years on eleven organic grain farms in Québec, Canada, to assess the impact of short crop rotations without forages (corn [Zea mays L.] -soybean [Glycine max (L.) Merr.] – small grain) on the earthworm community and soil health. Earthworm abundance and richness were generally low in both fields and field margins, with earthworm communities in cropped fields primarily dominated by endogeic species. Our findings suggest that long-term manure application, along with frequent use of cover crops, plus inversion tillage favours endogeic species in these systems. In contrast anecic and epigeic species increased under winter cereals. Legacy (prior three years) management practices were characterized using indices for crop diversity (CDI), soil tillage intensity rating (STIR) and organic amendment use (OAI). In-season management indices, in particular CDI (increased by cover crops and winter crops) and STIR, varied across crop fields and had a greater effect on soil health outcomes. A replicated potato (Solanum tuberculum L.) experiment in Prince Edward Island, Canada, compared the impact of contrasting three-year rotations including one full season cover crop on SOC dynamics and distribution. Treatments failed to differ in soil profile (0-45cm) SOC. Our results indicate that, carbon (C) inputs from full season cover crops were insufficient to offset SOC loss due to high tillage intensity in these rotations, affecting both topsoil and subsoil SOC stocks. Overall, this research has shown that reducing tillage intensity by increasing crop diversity was the main driver of earthworm community diversity and soil health. Rotations that enhance C inputs, such as via winter cover crops and cash crops residues, could buffer the earthworm community, and soil health and SOC, against the impacts of greater tillage intensity in short rotation cropping systems.