Protein aggregation is implicated in many human diseases. For example, Alzheimer’s disease is characterized by the misfolding of the amyloid beta (abeta) protein that disrupts cellular function by accumulating into fibrils with a characteristic cross-beta sheet structure. A wide variety of computational, sequence based tools are available to predict amyloidogenic “hot spots”; the regions in proteins that cause the formation of these toxic aggregates. Recently we have performed a comprehensive mutational analysis of abeta using a high throughput method termed Deep Mutational Scanning (DMS). Four hundred and fifty abeta variants were simultaneously assayed for solubility. An aggregation score that reflects the effect on the aggregation propensity of abeta was calculated for each variant. Using these data as a gold standard, we benchmarked aggregation prediction tools and assessed the ability of each algorithm to accurately predict aggregation. In particular, we were interested in identifying methodological weaknesses in these tools and then determining how to improve aggregation propensity prediction. I measured the correlation between several tools’ aggregation predictions for abeta single mutants with the measured aggregation scores. My initial analysis revealed that these tools are often insensitive and over-predict amyloidogenic regions, indicating an opportunity to improve these methods. Having comprehensively measured the relationship between abeta sequence and aggregation propensity, we are now in a position to assess and improve current aggregation prediction tools. Improving our ability to predict this critical process will enhance our understanding of diseases of protein aggregation.

Nitrous oxide (N₂O) is a pharmacologically active gas used clinically in dentistry and medicine. It is also an abused inhalant. Our lab has an established rat model of N₂O-induced hypothermia that can assess changes in core temperature (Tc) as well as the determinants of Tc, i.e., heat production (HP) and heat loss (HL). Drug-induced hypothermia elicits hypothermia-opposing autonomic HP responses that account for individual differences in the degree of hypothermia observed during an initial N₂O administration. Furthermore, these individual differences in HP also predict the rate of tolerance development at the level of Tc, predict an individual’s subsequent self-administration of N₂O, and predict the degree of allostatic over-correction of Tc that develops over repeated N₂O administrations. This study was designed to determine whether a rat’s HP response to a non-drug, cold challenge is correlated to its HP response during an initial N₂O administration. Individual rats (N=20) were tested in two different sessions in counterbalanced order; one session involved a 60% N₂O challenge while the rat was at 22°C ambient temperature and the other session involved a cold (10°C) ambient temperature challenge while the rat was exposed to control air. The N2O challenge was achieved as a rapid exponential rise to a steady-state 60% concentration whereas the cold challenge was a ramp change starting from 22°C. The two challenges were given 7 days apart. During each 2-h challenge, 1200 HP measurements were collected using indirect calorimetry and median values were evaluated. Results found a correlation (r=0.57; p=0.009) between HP responses during a N₂O challenge and HP responses during a cold challenge. These findings have translational importance because they suggest that individual differences related to a young individual’s addictive vulnerability may be predicted using a non-drug challenge which facilitates human research pertinent to understanding the pathogenesis and treatment of drug addiction.

American Crows (Corvus brachyrhynchos) form four different daily aggregations: after roosting at sunrise (post-roost aggregations), gathering or foraging during the day (diurnal activity centers), before roosting at sunset (pre-roost aggregations), and at night while sleeping (roosts). Crows vocalize in all these aggregation types, but the type of calls they make may differ between each context. We hypothesize that pre- and post-roost aggregations act as information centers where crows use calls to share information about potential foraging areas. We predict that this function would lead to differences in vocalizations between pre- and post-roost aggregations and diurnal activity centers. To address this prediction, we recorded vocalizations from crows in pre- and post-roost aggregations and diurnal activity centers within the greater Seattle area. We analyzed spectograms of these recordings and compared vocal patterns between aggregation types. The total number of crows varied by context, with more crows observed in post-roost aggregations than in diurnal activity centers. Controlling for the number of crows, the average duration of gaps between syllables differed across contexts and there was a trend for the number of calls per recording to differ. These variations in vocalizations may indicate that crows share different information across the three different contexts. In order to determine whether crows might be sharing information about foraging locations on pre- and post-roost aggregations, future work will incorporate playback studies to determine if calls given primarily in pre- and post-roost aggregations cause crows to change their behavior on diurnal activity centers. 

Crows are intelligent social animals, and many studies have focused on understanding their complex vocalizations. Here we examine the ways in which non-breeding crows use both vocal and non-vocal behaviors across contexts. Both vocal and non-vocal data were collected in the field using video camera recordings at three different behavioral contexts: post-roost aggregations, diurnal activity centers, and pre-roost aggregations. An ethogram of behaviors and vocalizations was developed from the video data and each behavior was illustrated. Each illustration diagramed the movement of the behaviors to provide a visual representation of findings to support future research. We found that cawing was linked with two non-vocal behaviors: head up and body down. This study provides insights into a more complex mode of communication in crows and how such communication differs across behavioral contexts.