Predicting Alzheimer's Disease Progression Using a Versatile Sequence-Length-Adaptive Encoder-Decoder LSTM Architecture.

Detecting Alzheimer's disease (AD) accurately at an early stage is critical for planning and implementing disease-modifying treatments that can help prevent the progression to severe stages of the disease. In the existing literature, diagnostic test scores and clinical status have been provided for specific time points, and predicting the disease progression poses a significant challenge. However, few studies focus on longitudinal data to build deep-learning models for AD detection. These models are not stable to be relied upon in real medical settings due to a lack of adaptive training and testing. We aim to predict the individual's diagnostic status for the next six years in an adaptive manner where prediction performance improves with the number of patient visits. This study presents a Sequence-Length Adaptive Encoder-Decoder Long Short-Term Memory (SLA-ED LSTM) deep-learning model on longitudinal data obtained from the Alzheimer's Disease Neuroimaging Initiative archive. In the suggested approach, decoder LSTM dynamically adjusts to accommodate variations in training sequence length and inference length rather than being constrained to a fixed length. We evaluated the model performance for various sequence lengths and found that for inference length one, sequence length nine gives the highest average test accuracy and area under the receiver operating characteristic curves of 0.920 and 0.982, respectively. This insight suggests that data from nine visits effectively captures meaningful cognitive status changes and is adequate for accurate model training. We conducted a comparative analysis of the proposed model against state-of-the-art methods, revealing a significant improvement in disease progression prediction over the previous methods.

Accurate Prediction of Alzheimer's Disease Progression Trajectory via a Novel Encoder-Decoder LSTM Architecture.

With an increase in life expectancy, there has been an increase in the aged population globally, and around 10% of this population suffers from Alzheimer's disease. Alzheimer's hugely impacts the quality of life and well-being of older adults and their caregivers. Thus, it is an emerging challenge to improve the early diagnosis and prognosis of the disease. Detecting hidden patterns in complex multidimensional datasets using recent advancements in machine learning provides a tremendous opportunity to meet this crucial need. In this study, using multimodal features and an individual's clinical status on one or more time points, we aimed to predict the individual's cognitive test scores, changes in Magnetic Resonance Imaging features, and the individual's diagnostic status for the next three years. We presented a novel Encoder-Decoder Long Short-Term Memory deep-learning model architecture for implementing our prediction. We applied it to data from the Alzheimer's Disease Neuroimaging Initiative, comprising longitudinal data of 1737 participants and 12,741 instances. The proposed model was found to be competent, with a validation accuracy of 0.941, a multi-class area under the curve of 0.960, and a test accuracy of 0.88 in identifying the various stages of Alzheimer's disease progression in patients with an initially cognitively normal or mild cognitive impairment which is a significant improvement over previous methods.Clinical relevance- The proposed approach can help improve diagnostic understanding of Alzheimer's Disease progression and assist in the early detection of various stages of Alzheimer's Disease based on clinical heterogeneity.

3-amino-7-phthalazinylbenzoisoxazoles as a novel class of potent, selective, and orally available inhibitors of p38alpha mitogen-activated protein kinase.

The p38 mitogen-activated protein kinase (MAPK) is a central signaling molecule in many proinflammatory pathways, regulating the cellular response to a multitude of external stimuli including heat, ultraviolet radiation, osmotic shock, and a variety of cytokines especially interleukin-1beta and tumor necrosis factor alpha. Thus, inhibitors of this enzyme are postulated to have significant therapeutic potential for the treatment of rheumatoid arthritis, inflammatory bowel disease, osteoporosis, and many other diseases where aberrant cytokine signaling is the driver of disease. Herein, we describe a novel class of 3-amino-7-phthalazinylbenzoisoxazole-based inhibitors. With relatively low molecular weight, these compounds are highly potent in enzyme and cell-based assays, with minimal protein shift in 50% human whole blood. Compound 3c was efficacious (ED 50 = 0.05 mg/kg) in the rat collagen induced arthritis (CIA) model.

Discovery of highly selective and potent p38 inhibitors based on a phthalazine scaffold.

Investigations into the structure-activity relationships (SAR) of a series of phthalazine-based inhibitors of p38 are described. These efforts originated from quinazoline 1 and through rational design led to the development of a series of orally bioavailable, potent, and selective inhibitors. Kinase selectivity was achieved by exploiting a collection of interactions with p38alpha including close contact to Ala157, occupation of the hydrophobic gatekeeper pocket, and a residue flip with Gly110. Substitutions on the phthalazine influenced the pharmacokinetic properties, of which compound 16 displayed the most desirable profile. Oral dosing (0.03 mg/kg) of 16 in rats 1 h prior to LPS challenge gave a >50% decrease in TNFalpha production.

Substituted aryl pyrimidines as potent and soluble TRPV1 antagonists.

Clinical candidate AMG 517 (1) is a potent antagonist toward multiple modes of activation of TRPV1; however, it suffers from poor solubility. Analogs with various substituents at the R region of 3 were prepared to improve the solubility while maintaining the potent TRPV1 activity of 1. Compounds were identified that maintained potency, had good pharmacokinetic properties, and improved solubility relative to 1.

Trisubstituted pyrimidines as transient receptor potential vanilloid 1 (TRPV1) antagonists with improved solubility.

A series of trisubstituted pyrimidines were synthesized to improve aqueous solubility of our first TRPV1 clinical candidate (1; AMG 517), while maintaining potent TRPV1 inhibitory activity. Structure-activity and structure-solubility studies led to the identification of compound 26. The aqueous solubility of 26 (>or=200microg/mL, 0.01 HCl; 6.7microg/mL, phosphate buffered saline (PBS); 150microg/mL, fasted-state simulated intestinal fluid (SIF)) was significantly improved over 1. In addition, compound 26 was found to be orally bioavailable (rat F(oral)=24%) and had potent TRPV1 antagonist activity (capsaicin IC(50)=1.5nM) comparable to that of 1.

Insight into catalysis of a unique GTPase reaction by a combined biochemical and FTIR approach.

Rap1 and Rap2 are the only small guanine nucleotide-binding proteins of the Ras superfamily that do not use glutamine for GTP hydrolysis. Moreover, Rap1GAP, which stimulates the GTPase reaction of Rap1 10(5)-fold, does not have the classical "arginine finger" like RasGAP but presumably, introduces an asparagine residue into the active site. Here, we address the requirements of this unique reaction in detail by combining various biochemical methods, such as fluorescence spectroscopy, stopped-flow and time-resolved Fourier transform infrared spectroscopy (FTIR). The fluorescence spectroscopic assay monitors primarily protein-protein interaction steps, while FTIR resolves simultaneously the elementary steps of functional groups labor-free, but it is less sensitive and needs higher concentrations. Combining both methods allows us to distinguish weather mechanistic defects caused by mutation are due to affinity or due to functionality. We show that several mutations of Asn290 block catalysis. Some of the mutants, however, still form a complex with Rap1*GDP in the presence of BeF(x) but not AlF(x), supporting the notion that fluoride complexes are indicators of the ground versus transition state. Mutational analysis also shows that Thr61 is not required for catalysis. While replacement of Thr61 of Rap1 by Leu eliminates GTPase activation by Rap1GAP, the T61A and T61Q mutants have only a minor effect on catalysis, but change the relative rates of cleavage and (P(i)(-)) release. While Rap1GAP(N290A) is completely inactive on wild-type Rap1, it can act on Rap1(T61Q), arguing that Asn290 in trans has a role in catalysis similar to that of the intrinsic Gln in Ras and Rho. Finally, since FTIR works at high, and thus mostly saturating, concentrations, it can clearly separate effects on affinity from purely catalytic modifications, showing that Arg388, conserved between RapGAPs and mutated in the homologous RheBGAP Tuberin, affects binding affinity severely but has no effect on the cleavage reaction itself.

Moving sound reduces arousal in psychosomatic patients.

The present study was designed to investigate the possible effect of moving sound stimulation on Galvanic Skin Response (GSR) in psychosomatic patients. Sixteen volunteers (psychosomatic = 8, healthy controls = 8) were exposed to a moving or stationary sound of same acoustic structure while GSR was monitored. Results indicated that moving sound produces a significant reduction of arousal in psychosomatic patients.

Discovery of potent, orally available vanilloid receptor-1 antagonists. Structure-activity relationship of N-aryl cinnamides.

The vanilloid receptor-1 (TRPV1 or VR1) is a member of the transient receptor potential (TRP) family of ion channels and plays a role in regulating the function of sensory nerves. A growing body of evidence demonstrates the therapeutic potential of TRPV1 modulators, particularly in the management of pain. As a result of our screening efforts, we identified (E)-3-(4-tert-butylphenyl)-N-(2,3-dihydrobenzo[b][1,4]dioxin-6-yl)acrylamide (1), an antagonist that blocks the capsaicin-induced and pH-induced uptake of (45)Ca(2+) in TRPV1-expressing Chinese hamster ovary cells with IC(50) values of 17 +/- 5 and 150 +/- 80 nM, respectively. In this report, we describe the synthesis and structure-activity relationship of a series of N-aryl cinnamides, the most potent of which (49a and 49b) exhibit good oral bioavailability in rats (F(oral) = 39% and 17%, respectively).

Fourier transform infrared spectroscopy on the Rap.RapGAP reaction, GTPase activation without an arginine finger.

GTPase activating proteins (GAPs) down-regulate Ras-like proteins by stimulating their GTP hydrolysis, and a malfunction of this reaction leads to disease formation. In most cases, the molecular mechanism of activation involves stabilization of a catalytic Gln and insertion of a catalytic Arg into the active site by GAP. Rap1 neither possesses a Gln nor does its cognate Rap-GAP employ an Arg. Recently it was proposed that RapGAP provides a catalytic Asn, which substitutes for the Gln found in all other Ras-like proteins (Daumke, O., Weyand, M., Chakrabarti, P. P., Vetter, I. R., and Wittinghofer, A. (2004) Nature 429, 197-201). Here, RapGAP-mediated activation has been investigated by time-resolved Fourier transform infrared spectroscopy. Although the intrinsic hydrolysis reactions of Rap and Ras are very similar, the GAP-catalyzed reaction shows unique features. RapGAP binding induces a GTP(*) conformation in which the three phosphate groups are oriented such that they are vibrationally coupled to each other, in contrast to what was seen in the intrinsic and the Ras.RasGAP reactions. However, the charge shift toward beta-phosphate observed with RasGAP was also observed for RapGAP. A GDP.P(i) intermediate accumulates in the GAP-catalyzed reaction, because the release of P(i) is eight times slower than the cleavage reaction, and significant GTP synthesis from GDP.P(i) was observed. Partial steps of the cleavage reaction are correlated with structural changes of protein side groups and backbone. Thus, the Rap.RapGAP catalytic machinery compensates for the absence of a cis-Gln by a trans-Asn and for the catalytic Arg by inducing a different GTP conformation that is more prone to be attacked by a water molecule.

The GTPase-activating protein Rap1GAP uses a catalytic asparagine.

Rap1 is a Ras-like guanine-nucleotide-binding protein (GNBP) that is involved in a variety of signal-transduction processes. It regulates integrin-mediated cell adhesion and might activate extracellular signal-regulated kinase. Like other Ras-like GNBPs, Rap1 is regulated by guanine-nucleotide-exchange factors (GEFs) and GTPase-activating proteins (GAPs). These GAPs increase the slow intrinsic GTPase reaction of Ras-like GNBPs by many orders of magnitude and allow tight regulation of signalling. The activation mechanism involves stabilization of the catalytic glutamine of the GNBP and, in most cases, the insertion of a catalytic arginine of GAP into the active site. Rap1 is a close homologue of Ras but does not possess the catalytic glutamine essential for GTP hydrolysis in all other Ras-like and Galpha proteins. Furthermore, RapGAPs are not related to other GAPs and apparently do not use a catalytic arginine residue. Here we present the crystal structure of the catalytic domain of the Rap1-specific Rap1GAP at 2.9 A. By mutational analysis, fluorescence titration and stopped-flow kinetic assay, we demonstrate that Rap1GAP provides a catalytic asparagine to stimulate GTP hydrolysis. Implications for the disease tuberous sclerosis are discussed.