Due to the limitations of small molecules in selectively and effectively targeting disease-causing genes, many human diseases remain without a cure. Disease-driving genes resistant to small molecule inhibition are now a potential target for PROTACs, organic compounds that engage both a target and a degradation-mediating E3 ligase, an approach showing great promise. However, the capacity of E3 ligases to process proteins for degradation is restricted and not universal. For the successful engineering of PROTACs, the degradation profile of a protein is of utmost importance. Despite this, only a limited number, around a few hundred, of proteins have been subjected to experimental testing for their compatibility with PROTACs. It is uncertain which other proteins within the entire human genome might be targeted by this PROTAC. We present PrePROTAC, a novel interpretable machine learning model that harnesses the power of protein language modeling in this paper. An external evaluation set, encompassing proteins from various gene families beyond those in the training data, yielded high accuracy for PrePROTAC, implying its generalizability across diverse protein types. Applying PrePROTAC to the human genome, we pinpoint over 600 understudied proteins that could react to PROTAC treatment. Three PROTAC compounds for novel drug targets involved in Alzheimer's disease are designed by us.
Evaluating in-vivo human biomechanics hinges on the accuracy of motion analysis. Although marker-based motion capture serves as the standard for analyzing human movement, its inherent lack of precision and practical challenges significantly circumscribe its usability in large-scale and real-world contexts. Overcoming these practical hindrances appears feasible through the implementation of markerless motion capture. However, the tool's ability to accurately determine joint motion and force characteristics has not been tested extensively across diverse human movements. During this study, 10 healthy subjects undertook 8 common daily tasks and exercise movements, and their motion data were captured using both marker-based and markerless methods concurrently. read more A comparative analysis using markerless and marker-based techniques was undertaken to determine the correlation (Rxy) and root-mean-square deviation (RMSD) in estimating ankle dorsi-plantarflexion, knee flexion, and the three-dimensional hip kinematics (angles) and kinetics (moments) during each movement. The estimations of ankle and knee joint angles and moments from markerless motion capture correlated well with those from marker-based systems, displaying a correlation coefficient (Rxy) of 0.877 for joint angles (RMSD 59) and 0.934 for moments (RMSD 266% height weight). The uniformity of high outcomes in markerless motion capture eases experimental complexity and allows for comprehensive analyses across broad samples. Hip angles and moments exhibited more disparity between the two systems (RMSD 67–159 and up to 715% height-weight ratios), especially during rapid movements like running. Markerless motion capture potentially improves the precision of hip-related data, yet further research is required to prove its reliability. read more The biomechanics community should persist in verifying, validating, and establishing best practices for markerless motion capture, which promises to significantly advance collaborative biomechanical research and enlarge the spectrum of real-world assessments required for clinical translation.
Manganese's duality exists in its essential nature for life processes and its toxicity at higher levels. read more Mutations in SLC30A10, first reported in 2012, were discovered as the inaugural inherited cause of elevated manganese levels. The apical membrane transport protein SLC30A10 transports manganese out of hepatocytes, into bile, and out of enterocytes, into the lumen of the gastrointestinal tract. A breakdown in the SLC30A10 protein's ability to regulate gastrointestinal manganese excretion causes a harmful buildup of manganese, leading to neurologic impairments, liver cirrhosis, polycythemia, and an overabundance of erythropoietin in the body. Neurologic and liver conditions are hypothesized to be a consequence of manganese toxicity. Polycythemia's association with excessive erythropoietin is well-established, but the basis of that excess in patients with SLC30A10 deficiency has yet to be characterized. We found that in Slc30a10-knockout mice, erythropoietin production is upregulated in the liver, while it is downregulated in the kidneys. Through the application of pharmacologic and genetic methods, we establish that the liver's expression of hypoxia-inducible factor 2 (Hif2), a transcription factor crucial for cellular adaptation to hypoxia, is essential for erythropoietin excess and polycythemia in Slc30a10-deficient mice, while hypoxia-inducible factor 1 (HIF1) has no significant impact. Livers deficient in Slc30a10 displayed altered gene expression, a majority associated with the cell cycle and metabolic functions, according to RNA sequencing analysis. Simultaneously, reduced hepatic Hif2 levels in these mice decreased the differential expression of almost half of the affected genes. In Hif2-dependent manner, hepcidin, a hormonal inhibitor of dietary iron absorption, is one of the genes downregulated in Slc30a10-deficient mice. Analyses of our data indicate that hepcidin's suppression elevates iron absorption, addressing the elevated erythropoiesis needs driven by an overabundance of erythropoietin. Finally, our findings also indicated that a reduction in hepatic Hif2 activity results in a decrease of manganese in tissues, despite the mechanism underlying this effect being presently unclear. Our investigation demonstrates that HIF2 is a vital driver of the pathophysiological features in cases of SLC30A10 deficiency.
In the general US adult population with hypertension, the predictive power of NT-proBNP has not been adequately characterized.
Using data from the 1999-2004 National Health and Nutrition Examination Survey, NT-proBNP measurements were taken for adults 20 years of age. In the adult population devoid of cardiovascular disease history, we evaluated the presence of elevated NT-pro-BNP levels stratified by blood pressure treatment and control categories. We evaluated the predictive capacity of NT-proBNP for mortality risk, across blood pressure treatment and control categories.
In the US adult population without CVD and with elevated NT-proBNP (a125 pg/ml), the prevalence of untreated hypertension was 62 million, that of treated and controlled hypertension 46 million, and that of treated but uncontrolled hypertension 54 million. After adjusting for factors including age, sex, BMI, and race/ethnicity, those with treated and controlled hypertension and elevated levels of NT-proBNP had a substantially higher risk of mortality from all causes (hazard ratio [HR] 229, 95% confidence interval [CI] 179-295) and cardiovascular mortality (HR 383, 95% CI 234-629) compared to those without hypertension and with low NT-proBNP (<125 pg/ml). For individuals taking antihypertensive medication, a systolic blood pressure (SBP) of 130-139 mm Hg coupled with elevated levels of NT-proBNP was associated with a higher risk of death from any cause compared to those with SBP below 120 mm Hg and low NT-proBNP levels.
Among adults without pre-existing cardiovascular conditions, NT-proBNP offers supplementary prognostic value, categorized by blood pressure classifications. Measurement of NT-proBNP holds potential for enhancing clinical hypertension treatment protocols.
In a population of adults free of cardiovascular disease, NT-proBNP can add to the prognostic understanding of blood pressure categories. Potential exists for optimizing hypertension treatment through the clinical application of NT-proBNP measurement.
A subjective memory of repeated passive and innocuous experiences, a consequence of familiarity, diminishes neural and behavioral responsiveness, while concurrently amplifying the recognition of new and distinct stimuli. A deeper understanding of the neural underpinnings of familiarity's internal model, and the cellular processes responsible for heightened novelty detection after repeated, passive exposure over multiple days, is still needed. We scrutinize the impact of repeated, passive exposure to an orientation-grating stimulus over multiple days on the spontaneous and non-familiar stimuli-evoked activity in neurons tuned to either familiar or non-familiar stimuli within the mouse visual cortex. Analysis revealed that familiarity engendered stimulus competition, which manifests as a decrease in stimulus selectivity in neurons tuned to familiar stimuli, contrasted with a concomitant enhancement in selectivity of neurons attuned to novel stimuli. Dominance in local functional connectivity is consistently exhibited by neurons attuned to novel stimuli. Beyond that, neurons that experience stimulus competition display a nuanced enhancement in responsiveness to natural images, which involve both familiar and unfamiliar orientations. The similarity between the responses to familiar grating stimuli and spontaneous activity increases is also demonstrated, signifying the presence of an internal model of modified experience.
Brain-computer interfaces (BCIs) using EEG technology, non-invasively, aim to replace or restore motor functions in patients with impairments, and offer direct brain-to-device communication to the general population. Amongst BCI paradigms, motor imagery stands out as a frequently utilized method; however, its performance varies considerably between users, and extensive training is often needed for effective control. Our proposed approach in this study involves a simultaneous integration of the MI and recently introduced Overt Spatial Attention (OSA) paradigms for the purpose of achieving BCI control.
Fifty BCI sessions, spanning five, were employed to assess the skill of 25 human subjects in maneuvering a virtual cursor across either one or two-dimensional spaces. The subjects utilized five diverse BCI protocols: MI used independently, OSA used independently, simultaneous MI and OSA targeting the same goal (MI+OSA), MI controlling one axis while OSA controlled the other (MI/OSA and OSA/MI), and the combined usage of MI and OSA.
Our findings indicate that the MI+OSA approach achieved the highest average online performance in 2D tasks, with a 49% Percent Valid Correct (PVC) rate, significantly surpassing the 42% PVC of MI alone, and exceeding, though not statistically, the 45% PVC of OSA alone.