Ionotropic Receptor-dependent cool cells control the transition of temperature preference in Drosophila larvae.

Temperature sensation guides animals to avoid temperature extremes and to seek their optimal temperatures. The larval stage of Drosophila development has a dramatic effect on temperature preference. While early-stage Drosophila larvae pursue a warm temperature, late-stage larvae seek a significantly lower temperature. Previous studies suggest that this transition depends on multiple rhodopsins at the late larval stage. Here, we show that early-stage larvae, in which dorsal organ cool cells (DOCCs) are functionally blocked, exhibit similar cool preference to that of wild type late-stage larvae. The molecular thermoreceptors in DOCCs are formed by three members of the Ionotropic Receptor (IR) family, IR21a, IR93a, and IR25a. Early-stage larvae of each Ir mutant pursue a cool temperature, similar to that of wild type late-stage larvae. At the late larval stage, DOCCs express decreased IR proteins and exhibit reduced cool responses. Importantly, late-stage larvae that overexpress IR21a, IR93a, and IR25a in DOCCs exhibit similar warm preference to that of wild type early-stage larvae. These data suggest that IR21a, IR93a, and IR25a in DOCCs navigate early-stage larvae to avoid cool temperatures and the reduction of these IR proteins in DOCCs results in animals remaining in cool regions during the late larval stage. Together with previous studies, we conclude that multiple temperature-sensing systems are regulated for the transition of temperature preference in fruit fly larvae.

Responses of different Drosophila species to temperature changes.

Temperature is a critical environmental variable that affects the distribution, survival and reproduction of most animals. Although temperature receptors have been identified in many animals, how these receptors respond to temperature is still unclear. Here, we describe an automated tracking method for studying the thermotactic behaviors of Drosophila larvae and adults. We built optimal experimental setups to capture behavioral recordings and analyzed them using free software, Fiji and TrackMate, which do not require programming knowledge. Then, we applied the adult thermotactic two-choice assay to examine the movement and temperature preferences of nine Drosophila species. The ability or inclination to move varied among these species and at different temperatures. Distinct species preferred various ranges of temperatures. Wild-type D. melanogaster flies avoided the warmer temperature in the warm avoidance assay and the cooler temperature in the cool avoidance assay. Conversely, D. bipectinata and D. yakuba did not avoid warm or cool temperatures in the respective assays, and D. biarmipes and D. mojavensis did not avoid the warm temperature in the warm avoidance assay. These results demonstrate that Drosophila species have different mobilities and temperature preferences, which will benefit further research in exploring molecular mechanisms of temperature responsiveness.

U-net model for brain extraction: Trained on humans for transfer to non-human primates.

Brain extraction (a.k.a. skull stripping) is a fundamental step in the neuroimaging pipeline as it can affect the accuracy of downstream preprocess such as image registration, tissue classification, etc. Most brain extraction tools have been designed for and applied to human data and are often challenged by non-human primates (NHP) data. Amongst recent attempts to improve performance on NHP data, deep learning models appear to outperform the traditional tools. However, given the minimal sample size of most NHP studies and notable variations in data quality, the deep learning models are very rarely applied to multi-site samples in NHP imaging. To overcome this challenge, we used a transfer-learning framework that leverages a large human imaging dataset to pretrain a convolutional neural network (i.e. U-Net Model), and then transferred this to NHP data using a small NHP training sample. The resulting transfer-learning model converged faster and achieved more accurate performance than a similar U-Net Model trained exclusively on NHP samples. We improved the generalizability of the model by upgrading the transfer-learned model using additional training datasets from multiple research sites in the Primate Data-Exchange (PRIME-DE) consortium. Our final model outperformed brain extraction routines from popular MRI packages (AFNI, FSL, and FreeSurfer) across a heterogeneous sample from multiple sites in the PRIME-DE with less computational cost (20 s~10 min). We also demonstrated the transfer-learning process enables the macaque model to be updated for use with scans from chimpanzees, marmosets, and other mammals (e.g. pig). Our model, code, and the skull-stripped mask repository of 136 macaque monkeys are publicly available for unrestricted use by the neuroimaging community at https://github.com/HumanBrainED/NHP-BrainExtraction.

Sliding Window INteraction Grammar (SWING): a generalized interaction language model for peptide and protein interactions.

The explosion of sequence data has allowed the rapid growth of protein language models (pLMs). pLMs have now been employed in many frameworks including variant-effect and peptide-specificity prediction. Traditionally, for protein-protein or peptide-protein interactions (PPIs), corresponding sequences are either co-embedded followed by post-hoc integration or the sequences are concatenated prior to embedding. Interestingly, no method utilizes a language representation of the interaction itself. We developed an interaction LM (iLM), which uses a novel language to represent interactions between protein/peptide sequences. Sliding Window Interaction Grammar (SWING) leverages differences in amino acid properties to generate an interaction vocabulary. This vocabulary is the input into a LM followed by a supervised prediction step where the LM's representations are used as features. SWING was first applied to predicting peptide:MHC (pMHC) interactions. SWING was not only successful at generating Class I and Class II models that have comparable prediction to state-of-the-art approaches, but the unique Mixed Class model was also successful at jointly predicting both classes. Further, the SWING model trained only on Class I alleles was predictive for Class II, a complex prediction task not attempted by any existing approach. For de novo data, using only Class I or Class II data, SWING also accurately predicted Class II pMHC interactions in murine models of SLE (MRL/lpr model) and T1D (NOD model), that were validated experimentally. To further evaluate SWING's generalizability, we tested its ability to predict the disruption of specific protein-protein interactions by missense mutations. Although modern methods like AlphaMissense and ESM1b can predict interfaces and variant effects/pathogenicity per mutation, they are unable to predict interaction-specific disruptions. SWING was successful at accurately predicting the impact of both Mendelian mutations and population variants on PPIs. This is the first generalizable approach that can accurately predict interaction-specific disruptions by missense mutations with only sequence information. Overall, SWING is a first-in-class generalizable zero-shot iLM that learns the language of PPIs.

TACI: An ImageJ Plugin for 3D Calcium Imaging Analysis.

Research in neuroscience has evolved to use complex imaging and computational tools to extract comprehensive information from data sets. Calcium imaging is a widely used technique that requires sophisticated software to obtain reliable results, but many laboratories struggle to adopt computational methods when updating protocols to meet modern standards. Difficulties arise due to a lack of programming knowledge and paywalls for software. In addition, cells of interest display movements in all directions during calcium imaging. Many approaches have been developed to correct the motion in the lateral (x/y) direction. This paper describes a workflow using a new ImageJ plugin, TrackMate Analysis of Calcium Imaging (TACI), to examine motion on the z-axis in 3D calcium imaging. This software identifies the maximum fluorescence value from all the z-positions a neuron appears in and uses it to represent the neuron's intensity at the corresponding t-position. Therefore, this tool can separate neurons overlapping in the lateral (x/y) direction but appearing on distinct z-planes. As an ImageJ plugin, TACI is a user-friendly, open-source computational tool for 3D calcium imaging analysis. We validated this workflow using fly larval thermosensitive neurons that displayed movements in all directions during temperature fluctuation and a 3D calcium imaging dataset acquired from the fly brain.

Cool and warm ionotropic receptors control multiple thermotaxes in Drosophila larvae.

Animals are continuously confronted with different rates of temperature variation. The mechanism underlying how temperature-sensing systems detect and respond to fast and slow temperature changes is not fully understood in fly larvae. Here, we applied two-choice behavioral assays to mimic fast temperature variations and a gradient assay to model slow temperature changes. Previous research indicates that Rhodopsin 1 (Rh1) and its phospholipase C (PLC) cascade regulate fast and slow temperature responses. We focused on the ionotropic receptors (IRs) expressed in dorsal organ ganglions (DOG), in which dorsal organ cool-activated cells (DOCCs) and warm-activated cells (DOWCs) rely on IR-formed cool and warm receptors to respond to temperature changes. In two-choice assays, both cool and warm IRs are sufficient for selecting 18°C between 18°C and 25°C but neither function in cool preferences between 25°C and 32°C. The Rh1 pathway, on the other hand, contributes to choosing preferred temperatures in both assays. In a gradient assay, cool and warm IR receptors exert opposite effects to guide animals to ∼25°C. Cool IRs drive animals to avoid cool temperatures, whereas warm IRs guide them to leave warm regions. The Rh1 cascade and warm IRs may function in the same pathway to drive warm avoidance in gradient assays. Moreover, IR92a is not expressed in temperature-responsive neurons but regulates the activation of DOWCs and the deactivation of DOCCs. Together with previous studies, we conclude that multiple thermosensory systems, in various collaborative ways, help larvae to make their optimal choices in response to different rates of temperature change.