Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Generative models”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,045 records · Page 58Linked to original sources
PubMed2004

A simplified model for V-ATPase H+ extrusion.

An analytical model of V-type H+-translocating ATPase (V-ATPase) was developed based on an approximation to the mechanochemical model of Grabe et al. (Biophys. J., pp. 2798-2813, vol. 78, 2000). Grabe's work utilizes structural information and physiological assumptions to construct a detailed mechanochemical model of the V-ATPase. Due to the complexity of their model, it does not give a readily usable mathematical expression for the V-ATPase current. Based on their analysis of the structure of the proton pump, we develop a two-compartment model of the V-ATPase, which contains a membrane "half-channel" for proton translocation separated by a hydrophilic strip and a hydrophobic wall from the cytoplasm. Using the Langevin equation to describe proton transport across the membrane, we simplify the model based on their assumptions on the molecular structure of the pump and arrive at a general form of solution to the proton pump flux driven by ATP hydrolysis based on assumptions on the physiological properties of the strip and the wall, as well as the two fluid compartments. In this process of simplification, we explicitly relate V-ATPase structure, stoichiometry, pump efficiency, and ATP hydrolysis energy to the active pump current. The simplified model is used to provide model-generated approximations to measured data from a variety of laboratories. In addition, it provides a very compact characterization of V-ATPase, which can be used as a proton extruder in a variety of different cell membranes, as well as in the membranes of intracellular organelles. Index Terms-Electrophysiology, mechanochemstry, molecular motors, proton extrusion

Biological Transport, Active↗
PubMed2001

Three-dimensional tissue assemblies: novel models for the study of Salmonella enterica serovar Typhimurium pathogenesis.

The lack of readily available experimental systems has limited knowledge pertaining to the development of Salmonella-induced gastroenteritis and diarrheal disease in humans. We used a novel low-shear stress cell culture system developed at the National Aeronautics and Space Administration in conjunction with cultivation of three-dimensional (3-D) aggregates of human intestinal tissue to study the infectivity of Salmonella enterica serovar Typhimurium for human intestinal epithelium. Immunohistochemical characterization and microscopic analysis of 3-D aggregates of the human intestinal epithelial cell line Int-407 revealed that the 3-D cells more accurately modeled human in vivo differentiated tissues than did conventional monolayer cultures of the same cells. Results from infectivity studies showed that Salmonella established infection of the 3-D cells in a much different manner than that observed for monolayers. Following the same time course of infection with Salmonella, 3-D Int-407 cells displayed minimal loss of structural integrity compared to that of Int-407 monolayers. Furthermore, Salmonella exhibited significantly lower abilities to adhere to, invade, and induce apoptosis of 3-D Int-407 cells than it did for infected Int-407 monolayers. Analysis of cytokine expression profiles of 3-D Int-407 cells and monolayers following infection with Salmonella revealed significant differences in expression of interleukin 1alpha (IL-1alpha), IL-1beta, IL-6, IL-1Ra, and tumor necrosis factor alpha mRNAs between the two cultures. In addition, uninfected 3-D Int-407 cells constitutively expressed higher levels of transforming growth factor beta1 mRNA and prostaglandin E2 than did uninfected Int-407 monolayers. By more accurately modeling many aspects of human in vivo tissues, the 3-D intestinal cell model generated in this study offers a novel approach for studying microbial infectivity from the perspective of the host-pathogen interaction.

Apoptosis↗
PubMed2003

Covalent linkage of the type-2 and type-3 structural mimics to model the active site structure of multicopper oxidases: synthesis and magneto- structural properties of two angular trinuclear copper(II) complexes.

Two new angular trinuclear copper(II) complexes of formulation [Cu(3)(HL)LL'](ClO(4)), where L' is imidazole (Him, 1) or 1-methylimidazole (1-MeIm, 2) and H(3)L is a Schiff base obtained from the condensation of salicylaldehyde and 1,3-diaminopropan-2-ol (2:1 mole ratio), are prepared from a reaction of [Cu(2)L(mu-Br)] and [Cu(HL)] in the presence of L' and isolated as perchlorate salts. The crystal structures of 1 and 2 consist of a trinuclear copper(II) unit formed by the covalent linkage of monomeric type-2 mimic and dimeric type-3 mimic precursor complexes to give an angular arrangement of the metal atoms in the core which is a model for the active site structure of blue multicopper oxidases. In 1 and 2, the coordination geometry of two terminal copper atoms is distorted square-planar. The central copper has a distorted square-pyramidal (4 + 1) geometry. The mean Cu...Cu distance is approximately 3.3 A. The complex has a diphenoxo-bridged dicopper(II) unit with the phenoxo oxygen atoms showing a planar geometry. In addition, the complex has an endogenous alkoxo-bridged dicopper(II) unit showing a pyramidal geometry for the oxygen atom. The 1:1 electrolytic complexes show a d-d band at 607 nm. Cyclic voltammetry of the complexes in MeCN containing 0.1 M TBAP using a glassy carbon working electrode displays a Cu(3)(II)/Cu(2)(II)Cu(I) couple near -1.0 V (vs SCE). The variable temperature magnetic susceptibility measurements in the range 300-18 K show antiferromagnetic coupling in the complexes giving magnetic moments of approximately 3.0 mu(B) at 300 K and approximately 2.1 mu(B) at 18 K for the tricopper(II) unit. The experimental susceptibility data are theoretically fitted using a model with Heisenberg spin-(1)/(2) Hamiltonian for a trimer of spin-(1)/(2) copper(II) ions having two exchange parameters involving the alkoxo-bridged dicopper(II) (J1) and the diphenoxo-bridged dicopper(II) (J2) units, giving J1 and J2 values of -82.7, -73 cm(-1) for 1 and -98.3, -46.1 cm(-1) for 2, respectively. The structural features indicate a higher magnitude of anitiferromagnetic coupling in the alkoxo-bridged unit based on the greater value of the Cu-O-Cu angle in comparison to the diphenoxo-bridged unit. The core structures of 1 and 2 compare well with the first generation model complexes for the active site structure of multicopper oxidases in the oxidized form. The crystal structure of 1 exhibits a lamellar structure with a gap of approximately 7 A containing water molecules in the interlamellar space. Complex 2 forms a hexanuclear species due to intermolecular hydrogen bonding interactions involving two trimeric units. The crystal packing diagram of 2 displays formation of a three-dimensional framework with cavities containing the perchlorate anions.

Algorithms↗
PubMed1992

The population genetics of alleles affecting enzyme activity.

It is possible to predict the population genetics of allozymes by assuming that fitness is proportional to flux through a biochemical pathway. The model presented here extends previous work by incorporating two additional features of biological realism. Firstly, that more than one biochemical route may exist between any two metabolites. The major routes have been identified as the classical biochemical pathways but in the event of a mutation blocking a major route, minor routes become significant. These minor routes are named "bypass fluxes" and have profound effects on the population genetics of allozymes. Secondly, recent work has suggested that a metabolic cost is associated with enzyme synthesis; this will constitute an additional selective pressure on alleles which affect the amount of enzyme synthesized. The model generates a fitness curve which predicts the fitness associated with any level of enzyme activity. It can utilize data on null or near-null, structural or regulatory, mutations in the presence or absence of bypass fluxes. When data from natural populations of Drosophila are investigated, it is concluded that selection pressures acting on enzyme variants may be much higher than previously thought.

Animals↗
PubMed2000

Purification, crystallization and preliminary crystallographic studies of an integral membrane protein, cytochrome bo3 ubiquinol oxidase from Escherichia coli.

Cytochrome bo(3) ubiquinol oxidase has been successfully purified for crystallization. Single crystals of this integral membrane protein diffract X-rays to 3.5 A resolution and belong to the orthorhombic space group C222(1). From the diffraction data, the unit-cell parameters were determined to be a = 91.3, b = 370.3, c = 232.4 A. The crystals have a solvent content of 59% and contain two molecules per asymmetric unit. A search model generated from the structures of cytochrome c oxidase from Paracoccus denitrificans and the extrinsic domain of cytochrome bo(3) ubiquinol oxidase from Escherichia coli was used for molecular-replacement studies, resulting in a solution with sensible molecular packing.

Crystallization↗
PubMed2006

The influence of phoneme position overlap on the phonemic similarity effect in nonword recall.

The current research examined the predictions that short-term memory models generate for the phonological similarity effect, when similarity was defined in different ways. Three serial recall experiments with consonant-vowel-consonant (CVC) nonwords are reported, where the position of the phonemes that list items shared was manipulated (i.e., shared vowel and final consonant [_VC; Experiment 1], initial consonant and vowel [CV_; Experiment 2], or the two consonants [C_C; Experiment 3]. The results show that the position of common phonemes in nonwords has differential effects on order and item information. The findings are discussed in relation to previous research into the effect of phonemic similarity on nonword recall, and modifications to current short-term memory models are proposed.

Adolescent↗
PubMed1991

Artificial neural networks for source localization in the human brain.

Source localization in the brain remains an ill-posed problem unless further constraints about the type of sources and the head model are imposed. Human head is modeled in various ways depending critically on the computing power available and/or the required level of accuracy. Sophisticated and truly representative models may yield more accurate results in general, but at the cost of prohibitively long computer times and huge memory requirements. In conventional source localization techniques, solution source parameters are taken as those which minimize an index of performance, defined relative to the model-generated and clinically measured voltages. We propose the use of a neural network in the place of commonly employed minimization algorithms such as the Simplex Method and the Marquardt algorithm, which are iterative and time consuming. With the aid of the error-backpropagation technique, a neural network is trained to compute source parameters, starting from a voltage set measured on the scalp. Here we describe the methods of training the neural network and investigate its localization accuracy. Based on the results of extensive studies, we conclude that neural networks are highly feasible as source localizers. A trained neural network's independence of localization speed from the head model, and the rapid localization ability, makes it possible to employ the most complex head model with the ease of the simplest model. No initial parameters need to be guessed in order to start the calculation, implying a possible automation of the entire localization process. One may train the network on experimental data, if available, thereby possibly doing away with head models.

Brain↗
PubMed2005

Gastric adenocarcinoma: review and considerations for future directions.

OBJECTIVE: This update reviews the epidemiology and surgical management, and the controversies of gastric adenocarcinoma. We provide the relevance of outcome data to surgical decision-making and discuss the application of gene-expression analysis to clinical practice. SUMMARY BACKGROUND DATA: Gastric cancer mortality rates have remained relatively unchanged over the past 30 years, and gastric cancer continues to be one of the leading causes of cancer-related death. Well-conducted studies have stimulated changes to surgical decision-making and technique. Microarray studies linked to predictive outcome models are poised to advance our understanding of the biologic behavior of gastric cancer and improve surgical management and outcome. METHODS: We performed a review of the English gastric adenocarcinoma medical literature (1980-2003). This review included epidemiology, pathology and staging, surgical management, issues and controversies in management, prognostic variables, and the application of outcome models to gastric cancer. The results of DNA microarray analysis in various cancers and its predictive abilities in gastric cancer are considered. RESULTS: Prognostic studies have provided valuable data to better the understanding of gastric cancer. These studies have contributed to improved surgical technique, more accurate pathologic characterization, and the identification of clinically useful prognostic markers. The application of microarray analysis linked to predictive models will provide a molecular understanding of the biology driving gastric cancer. CONCLUSIONS: Predictive models generate important information allowing a logical evolution in the surgical and pathologic understanding and therapy for gastric cancer. However, a greater understanding of the molecular changes associated with gastric cancer is needed to guide surgical and medical therapy.

Adenocarcinoma↗
PubMed2002

Intralysosomal cystine accumulation in mice lacking cystinosin, the protein defective in cystinosis.

Cystinosis is an autosomal recessive disorder characterized by an accumulation of intralysosomal cystine. The causative gene, CTNS, encodes cystinosin, a seven-transmembrane-domain protein, which we recently showed to be a lysosomal cystine transporter. The most severe and frequent form of cystinosis, the infantile form, appears around 6 to 12 months, with a proximal tubulopathy (de Toni-Debré-Fanconi syndrome) and ocular damage. End-stage renal failure is reached by 10 years of age. Accumulation of cystine in all tissues eventually leads to multisystemic disease. Treatment with cysteamine, which reduces the concentration of intracellular cystine, delays disease progression but has undesirable side effects. We report the first Ctns knockout mouse model generated using a promoter trap approach. We replaced the last four Ctns exons by an internal ribosome entry site-betagal-neo cassette and showed that the truncated protein was mislocalized and nonfunctional. Ctns(-/-) mice accumulated cystine in all organs tested, and cystine crystals, pathognomonic of cystinosis, were observed. Ctns(-/-) mice developed ocular changes similar to those observed in affected individuals, bone defects and behavioral anomalies. Interestingly, Ctns(-/-) mice did not develop signs of a proximal tubulopathy, or renal failure. A preliminary therapeutic trial using an oral administration of cysteamine was carried out and demonstrated the efficiency of this treatment for cystine clearance in Ctns(-/-) mice. This animal model will prove an invaluable and unique tool for testing emerging therapeutics for cystinosis.

Alleles↗
PubMed2006

Hypertension, kidney, and transgenics: a fresh perspective.

In this review, we outline the application and contribution of transgenic technology to establishing the genetic basis of blood pressure regulation and its dysfunction. Apart from a small number of examples where high blood pressure is the result of single gene mutation, essential hypertension is the sum of interactions between multiple environmental and genetic factors. Candidate genes can be identified by a variety of means including linkage analysis, quantitative trait locus analysis, association studies, and genome-wide scans. To test the validity of candidate genes, it is valuable to model hypertension in laboratory animals. Animal models generated through selective breeding strategies are often complex, and the underlying mechanism of hypertension is not clear. A complementary strategy has been the use of transgenic technology. Here one gene can be selectively, tissue specifically, or developmentally overexpressed, knocked down, or knocked out. Although resulting phenotypes may still be complicated, the underlying genetic perturbation is a starting point for identifying interactions that lead to hypertension. We recognize that the development and maintenance of hypertension may involve many systems including the vascular, cardiac, and central nervous systems. However, given the central role of the kidney in normal and abnormal blood pressure regulation, we intend to limit our review to models with a broadly renal perspective.

Animals↗
PubMed1990

Urea kinetics and clinical evaluation of the haemodialysis patient.

Urea kinetic modelling (UKM) was performed on 62 patients in a haemodialysis unit not normally using kinetic methods. Without knowledge of the results, four nephrologists, four nurses and the patients themselves evaluated adequacy of dialysis (eAD) and daily protein intake (eDPI). Thirty-two patients had Kt/V less than 1.0, and 17 patients had Kt/V less than 0.9. Estimated improvement of the efficacy of treatment after the intervention of a physician was minor. Seven patients had a protein catabolic rate (pcr) at less than 0.8 g/kg per day. On average physicians identified five of these. Both nurses and doctors exhibited highly significant correlations between Kt/V and eAD, and between pcr and eDPI, but the correlation coefficients were generally modest (typically below 0.4). When patients evaluated themselves, no significant correlations were found. Examined individually, all four physicians' decisions about eAD correlated better with model-generated decisions than with eAD stated by their colleagues. It is concluded that UKM should be used to secure adequate and more uniform treatment prescription. There is no 'clinical standard' competing with UKM. Nurses make satisfactory evaluations compared to doctors, but the patients are unable to assess the adequacy of their dialysis or diet.

Adult↗
PubMed2006

Nonlinearities in color coding: compensating color appearance for the eye's spectral sensitivity.

Most wavelengths change hue when mixed with white light. These changes, known as the Abney effect, have been extensively studied to characterize nonlinearities in the neural coding of color, but their potential function remains obscure. We measured the Abney effect in a new way--by varying the bandwidth of the spectrum rather than mixing with white--and this leads to a new interpretation of the role of nonlinear responses in color appearance. Because of the eye's limited spectral sensitivity, increasing the bandwidth of a spectrum changes the relative responses in the three classes of cone receptor and thus would change hue if the percept were tied to a fixed cone ratio. However, we found that hue is largely independent of bandwidth and thus constant for a constant peak wavelength for stimuli with Gaussian spectra. This suggests that color appearance is compensated for the eye's spectral filtering, and that this compensation embodies specific perceptual inferences about how natural spectra vary. When a wavelength is instead diluted with white light--which does not bias the cone ratios--then the same compensation predicts changes in hue because the "right" response is made to the "wrong" stimulus. This model generates constant hue loci that are qualitatively consistent with measures of the Abney effect and provides a novel functional account of such effects in color appearance, in which postreceptoral responses are adjusted so that constant hue percepts are tied to consistent physical properties of the environment rather than consistent physiological properties such as the cone ratios.

Adaptation, Physiological↗
PubMed1999

Three-dimensional pharmacophore hypotheses of octopamine receptor responsible for the inhibition of sex-pheromone production in Helicoverpa armigera.

Three-dimensional pharmacophore hypotheses were built on the basis of a set of nine octopamine (OA) agonists responsible for the inhibition of sex-pheromone production in Helicoverpa armigera. Of the 10 models generated by the program Catalyst/Hypo, hypotheses including hydrogen-bond acceptor (HBA), hydrophobic (Hp), and hydrophobic aliphatic (HpAl) features were considered important and predictive in evaluating OA agonists. An HBA and four hydrophobic features are the minimum components of an effective OA agonist-binding hypothesis, which resembles the results of binding activity to locust OAR3. Active agonists mapped well onto all of the features of the hypothesis, such as HBA, Hp, and HpAl features. On the other hand, inactive compounds lacking binding affinity were shown to be poorly capable of achieving an energetically favorable conformation shared by the active molecules in order to fit the 3D chemical feature pharmacophore models. Those hypotheses are considered useful in designing new leads for more active compounds. Further research on the comparison of models from agonists may help elucidate the mechanisms of OA receptor-ligand interactions.

Animals↗
PubMed2025

Embed-Search-Align: DNA sequence alignment using Transformer models.

MOTIVATION: DNA sequence alignment, an important genomic task, involves assigning short DNA reads to the most probable locations on an extensive reference genome. Conventional methods tackle this challenge in two steps: genome indexing followed by efficient search to locate likely positions for given reads. Building on the success of Large Language Models in encoding text into embeddings, where the distance metric captures semantic similarity, recent efforts have encoded DNA sequences into vectors using Transformers and have shown promising results in tasks involving classification of short DNA sequences. Performance at sequence classification tasks does not, however, guarantee sequence alignment, where it is necessary to conduct a genome-wide search to align every read successfully, a significantly longer-range task by comparison. RESULTS: We bridge this gap by developing a "Embed-Search-Align" (ESA) framework, where a novel Reference-Free DNA Embedding (RDE) Transformer model generates vector embeddings of reads and fragments of the reference in a shared vector space; read-fragment distance metric is then used as a surrogate for sequence similarity. ESA introduces: (i) Contrastive loss for self-supervised training of DNA sequence representations, facilitating rich reference-free, sequence-level embeddings, and (ii) a DNA vector store to enable search across fragments on a global scale. RDE is 99% accurate when aligning 250-length reads onto a human reference genome of 3 gigabases (single-haploid), rivaling conventional algorithmic sequence alignment methods such as Bowtie and BWA-Mem. RDE far exceeds the performance of six recent DNA-Transformer model baselines such as Nucleotide Transformer, Hyena-DNA, and shows task transfer across chromosomes and species. AVAILABILITY AND IMPLEMENTATION: Please see https://anonymous.4open.science/r/dna2vec-7E4E/readme.md.

Sequence Analysis, DNA↗
PubMed2006

A model for the drug release from a polymer matrix tablet--effects of swelling and dissolution.

A model for simulating the drug release from a swelling and dissolving polymer tablet is presented and verified to data. The model is based on a mechanistic approach, and it can therefore be employed to study the sensitivity of true physical constants, for instance the drug diffusion coefficient or the drug solubility. The model generates the drug and polymer release profiles and the front positions of the total tablet, the solid core, and of the solid-drug-solubilized-drug interface. The convective contribution to mass transfer is shown to be of great importance. This is most markedly noticed for slowly diffusing drugs. In a simulation with a low value of the drug diffusion coefficient, it is shown that the initial drug release rate is faster than the polymer dissolution rate, followed by a second stage with a slower drug release rate. Furthermore, it is shown that polymer dissolution influences the drug release profile significantly, but not the front position of saturated drug in the gel layer. The model is verified against drug release and polymer dissolution data for the slightly soluble drug Methyl paraben and the soluble drug Saligenin in a poly (ethylene oxide) tablet, resulting in good agreement between model and experiments.

Benzyl Alcohols↗
PubMed2003

Psychosocial consequences of dental fear and anxiety.

OBJECTIVES: The aim of this study was to examine the negative psychosocial impacts of dental anxiety in a sample of dentally fearful and anxious individuals recruited from the general population. The associations between psychosocial impacts, dental anxiety scale (DAS) scores and other severe fears were explored. METHODS: One hundred and thirty-five subjects who were anxious or fearful about dental treatment were divided into low and high general fear groups based on the number of other severe fears they reported. Negative psychosocial impacts were assessed using a modified form of the scale developed by Kent et al. (1996). This consisted of three dimensions: psychological reactions, social relationships and avoidance/inhibition. Other measures included self-ratings of oral, general and emotional health and scales to assess self-esteem and morale. RESULTS: Overall, 93.1% of subjects reported one or more impacts. Those in the high-fear group had higher psychosocial impact scores than those in the low-fear group (means of 4.19 vs. 2.85; P < 0.05). Differences were most marked with respect to psychological consequences and avoidance/inhibition. The high-fear group had scores indicative of lower self-esteem and lower morale. Forward stepwise linear and logistic regression analyses indicated that both dental anxiety and general fearfulness contributed to these negative outcomes. However, the latter was a more consistent predictor in that it entered six of seven models generated while the former entered only four. CONCLUSION: The study indicated that dental fear and anxiety have pervasive psychosocial consequences, and that these are more marked among subjects with high levels of general fearfulness. It also provided evidence of the validity of a modified form of the psychosocial impact scale developed by Kent et al. (1996).

Adult↗
PubMed1993

[Dielectric behavior of isolated rat submandibular glands: simulation with a vesicle-inclusion cell model].

In an attempt to correlate the passive electrical properties of the secretory tissue with its structure, I measured AC admittances for isolated rat submandibular glands, over the frequency range between 100 Hz and 500 MHz. Animals were divided into three groups: control and treatment with two different secretagogues, isoproterenol and pilocarpine. Dielectric dispersions observed from control glands had two characteristic frequencies at 2.6 kHz and 59 kHz; the former was of the alpha-type and the latter of the beta-type. Secretagogue-stimulated glands exhibited broader beta-dispersion curves. These dispersion data were analysed on the basis of a "two-shell model" with vesicle inclusions, in which two concentric shells represent the cell membrane and the nuclear envelope, and membrane-bounded vesicles are meant for secretory granules dispersed in the inter-shell space. Simulation by this model generated a good fit to data from the three groups. The analyses revealed that: (i) electric capacities for the cell and granular membranes were 2.1 and 0.9 microF/cm2, respectively; (ii) the conductivity of extracellular matrix was close to that of blood plasma; (iii) the conductivity ratio between cytoplasm and extracellular matrix was approx. 0.4; and (iv) the conductivity ratio between intragranular space and cytoplasm was approx. 0.7.

Animals↗
PubMed2000

3-(Hydroxymethyl)-bearing phosphatidylinositol ether lipid analogues and carbonate surrogates block PI3-K, Akt, and cancer cell growth.

Phosphatidylinositol 3-kinase (PI3-K) phosphorylates the 3-position of phosphatidylinositol to give rise to three signaling phospholipids. Binding of the pleckstrin homology (PH) domain of Akt to membrane PI(3)P's causes the translocation of Akt to the plasma membrane bringing it into contact with membrane-bound Akt kinase (PDK1 and 2), which phosphorylates and activates Akt. Akt inhibits apoptosis by phosphorylating Bad, thus promoting its binding to and blockade of the activity of the cell survival factor Bcl-x. Herein we present the synthesis and biological activity of several novel phosphatidylinositol analogues and demonstrate the ability of the carbonate group to function as a surrogate for the phosphate moiety. Due to a combination of their PI3-K and Akt inhibitory activities, the PI analogues 2, 3, and 5 proved to be good inhibitors of the growth of various cancer cell lines with IC(50) values in the 1-10 microM range. The enhanced Akt inhibitory activity of the axial hydroxymethyl-bearing analogue 5 compared to its equatorial counterpart 6 is rationalized based upon postulated differences in the H-bonding patterns of these compounds in complex with a homology modeling generated structure of the PH domain of Akt. This work represents the first attempt to examine the effects of 3-modified PI analogues on these two crucial cell signaling proteins, PI3-K and Akt, in an effort to better understand their cell growth inhibitory properties.

Antineoplastic Agents↗