Search PubMed⌕ Search

Biomedical subjects

M K Ali

Publications and source records attributed to M K Ali.

At least 19 recordsLinked to original sources
PubMed2003

Convergence of reinforcement learning algorithms and acceleration of learning.

The techniques of reinforcement learning have been gaining increasing popularity recently. However, the question of their convergence rate is still open. We consider the problem of choosing the learning steps alpha(n), and their relation with discount gamma and exploration degree epsilon. Appropriate choices of these parameters may drastically influence the convergence rate of the techniques. From analytical examples, we conjecture optimal values of alpha(n) and then use numerical examples to verify our conjectures.

Journal Article↗
PubMed2002

Neural networks for estimating intrinsic dimension.

We consider the problem of feature extraction and determination of intrinsic dimensionality of observation data. One of the common approaches to this problem is to use autoassociative neural networks with a "bottleneck" projecting layer. We propose a different approach in which a neural network performs a topological mapping that creates a nonlinear lower-dimensional projection of the data. The mapping preserves relative distances of neighbors. This technique can be efficiently implemented with the help of radial basis function networks, and it is significantly faster than training an autoassotiative network. We show that the proposed technique can be used for estimating the dimension of minimal mathematical model from time series data.

Journal Article↗
PubMed2001

Q learning in the minority game.

We present a numerical investigation of the minority game model, where the dynamics of the agents is described by the Q-learning algorithm. The numerical results show that the Q-learning dynamics is suppressing the "crowd effect," which is characteristic of the minority game with standard inductive dynamics, and it converges to a stationary state that is close to the optimal Nash equilibrium of the game.

Journal Article↗
PubMed2001

Robust chaos in smooth unimodal maps.

Robust chaos is defined by the absence of periodic windows and coexisting attractors in some neighborhood of the parameter space. It has been conjectured that robust chaos cannot occur in smooth systems [E. Barreto, B. Hunt, and C. Grebogi, Phys. Rev. Lett. 78, 4561 (1997); 80, 3049 (1998)]. Contrary to this conjecture, we describe a general procedure for generating robust chaos in smooth unimodal maps.

Journal Article↗
PubMed2001

The multidomain xylanase Xyn10B as a cellulose-binding protein in Clostridium stercorarium.

The cells of Clostridium stercorarium F-9 grown on cellobiose bound to insoluble cellulose allomorphs such as phosphoric acid-swollen cellulose (ASC). Treatment of the cells with 3 M guanidine hydrochloride extracted surface-layer proteins from the cells and abolished the affinity of the cells for ASC. SDS-polyacrylamide gel electrophoresis, zymogram, and immunological analyses indicated that one of the major surface layer proteins was Xyn10B, which is a modular xylanase comprising two family 22 carbohydrate-binding modules (CBMs), a family 10 catalytic domain of glycosyl hydrolases, a family 9 CBM, and two S-layer homologous (SLH) domains. The C. stercorarium F-9 cells treated with guanidine hydrochloride coprecipitated with ASC upon the addition of a derivative of Xyn10B containing both a CBM and SLH domain in addition to a catalytic domain, but not a derivative without Xyn10B-SLH domains, suggesting that Xyn10B functioned as a cellulose-binding protein in C. stercorarium F-9.

Adsorption↗
PubMed2001

Chaotic neural control.

We consider the problem of stabilizing unstable equilibria by discrete controls (the controls take discrete values at discrete moments of time). We prove that discrete control typically creates a chaotic attractor in the vicinity of an equilibrium. Artificial neural networks with reinforcement learning are known to be able to learn such a control scheme. We consider examples of such systems, discuss some details of implementing the reinforcement learning to controlling unstable equilibria, and show that the arising dynamics is characterized by positive Lyapunov exponents, and hence is chaotic. This chaos can be observed both in the controlled system and in the activity patterns of the controller.

Journal Article↗
PubMed2001

Critical transition in the constrained traveling salesman problem.

We investigate the finite size scaling of the mean optimal tour length as a function of density of obstacles in a constrained variant of the traveling salesman problem (TSP). The computational experience pointed out a critical transition (at rho(c) approximately 85%) in the dependence between the excess of the mean optimal tour length over the Held-Karp lower bound and the density of obstacles.

Journal Article↗
PubMed2001

Effect of phospholipase C, trypsin and neuraminidase on binding of bilirubin to mammalian erythrocyte membranes.

Binding of bilirubin to erythrocyte membranes of human, buffalo, sheep and goat was studied after phospholipase C, trypsin and neuraminidase treatment. Phospholipase C and trypsin treatment of membranes greatly enhanced the bilirubin binding in all mammalian species, whereas, neuraminidase treatment resulted into a small increase in the membrane-bound bilirubin. Human erythrocyte membranes bound the highest amount of bilirubin, whereas buffalo, sheep and goat erythrocyte membranes showed different mode of bilirubin binding. The order of bilirubin binding to unmodified as well as neuraminidase-treated erythrocyte membranes was: human>sheep>buffalo>goat; the order was: human>buffalo>sheep>goat; in phospholipase C- and trypsin-treated erythrocyte membranes. These binding results indicate that membrane phospholipids are directly involved in the interaction of bilirubin with the membranes as the differences observed in the membrane-bound bilirubin among mammalian species were directly correlated with the sum of choline phospholipids, especially phosphatidylcholine and sphingomyelin content of the erythrocyte membranes. The negatively charged phosphate moiety of phospholipids of the membranes appears to inhibit a large amount of bilirubin binding to the membrane as its removal by phospholipase C greatly enhanced the binding. Furthermore, membrane proteins and carbohydrate also seem to play a significant regulatory function on the binding as their degradation and/or removal in the form of glycopeptides by trypsin expose a large number of bilirubin binding sites.

Animals↗
PubMed2001

Validation in the canine model of a new non-invasive method of measuring coronary blood flow reserve: split dose thallium-201 rest/stress imaging.

Diffusely impaired coronary blood flow reserve is difficult to measure non-invasively. We developed and tested a quantitative non-invasive method of measuring coronary blood flow reserve using thallium-201 perfusion imaging. Ten anesthetized dogs were injected simultaneously at rest with thallium-201 and either Ru-103 or Sn-113 microspheres. SPECT images were obtained followed by varying doses of intravenous adenosine, and a second thallium-201 dose was injected simultaneously with either Nb-95 or Sc-46 microspheres. SPECT images were then repeated. The heart was removed, sectioned and counted, along with arterial blood samples. Blood flow was calculated at rest and stress. Peak resting counts in four regions in each of three SPECT slices were subtracted from stress values and stress/rest thallium-201 count ratios (coronary flow reserve (CFR)) were calculated and correlated with the corresponding microsphere flow ratios. Overall correlation of the imaging and microsphere flow ratios was 0.77 (p = 0.0001). Regional correlation coefficients ranged from 0.65-0.86 (p = 0.0001). Coronary blood flow reserve ratios by the microsphere method ranged from 0.7 to 5.3, and by thallium-201 imaging from 0.33-2.45. The non-invasively measured coronary blood flow reserve with thallium-201 imaging and adenosine stress correlates well with microsphere-measured coronary blood flow reserve over a wide range of coronary flows, and should be useful in clinical studies of CFR impairment.

Adenosine↗
PubMed2001

Importance of the carbohydrate-binding module of Clostridium stercorarium Xyn10B to xylan hydrolysis.

The Clostridium stercorarium xylanase Xyn10B is a modular enzyme comprising two thermostabilizing domains, a family 10 catalytic domain of glycosyl hydrolases, a family 9 carbohydrate-binding module (CBM), and two S-layer homologous (SLH) domains [Biosci. Biotechnol. Biochem., 63, 1596-1604 (1999)]. To investigate the role of this CBM, we constructed two derivatives of Xyn10B and compared their hydrolytic activity toward xylan and some preparations of plant cell walls; Xyn10BdeltaCBM consists of a catalytic domain only, and Xyn10B-CBM comprises a catalytic domain and a CBM. Xyn10B-CBM bound to various insoluble polysaccharides including Avicel, acid-swollen cellulose, ball-milled chitin, Sephadex G-25, and amylose-resin. A cellulose binding assay in the presence of soluble saccharides suggested that the CBM of Xyn10B had an affinity for even monosaccharides such as glucose, galactose, xylose, mannose and ribose. Removal of the CBM from the enzyme negated its cellulose- and xylan-binding abilities and severely reduced its enzyme activity toward insoluble xylan and plant cell walls but not soluble xylan. These findings clearly indicated that the CBM of Xyn10B is important in the hydrolysis of insoluble xylan. This is the first report of a family 9 CBM with an affinity for insoluble xylan in addition to crystalline cellulose and the ability to increase hydrolytic activity toward insoluble xylan.

Catalysis↗
PubMed2001

Effect of metal ions on binding of bilirubin to erythrocyte membranes.

Binding of bilirubin to different erythrocyte membranes, namely, human, buffalo, sheep and goat, pre-incubated with different concentrations of metal ions was studied. The results showed that among the different metal ions used, Ca2+ had the highest potential in increasing the amount of bound bilirubin followed by Sr2+ and Mg2+, whereas Ba2+ had the lowest potential. Treatment of these membranes with Ca2+ led to an increase in the amount of bound bilirubin in all membranes. However, human erythrocyte membranes pretreated with Ca2+, bound the highest amount of bilirubin compared to other erythrocyte membranes. Increase in bilirubin binding upon Ca2+-treatment can be ascribed to shielding effect, redistribution of phospholipids as well as increase in surface hydrophobicity induced by Ca2+.

Animals↗
PubMed2000

Differential accessibility of bilirubin to erythrocyte membrane vesicles bearing different structural features.

Interaction of bilirubin with different types of erythrocyte membrane vesicles such as unsealed, heterogeneous, sealed and inside-out membrane vesicles prepared from human and goat erythrocytes was studied. Out of various types of membrane vesicles, in both species, unsealed membrane vesicles bound quantitatively higher amounts of bilirubin followed by heterogeneous and sealed membrane vesicles whereas inside-out membrane vesicles bound the lowest amount of bilirubin. These differences in the amount of bound bilirubin to different membrane vesicles were correlated well with the percentage accessibility of sialic acid to neuraminidase in these membranes suggesting that bilirubin bound preferentially to the outer layer of erythrocyte membranes than the inner layer. Further, membrane vesicles prepared from human erythrocytes bound higher amounts of bilirubin than those prepared from goat erythrocytes. This can be ascribed to different phospholipid composition of these membranes.

Animals↗
PubMed2000

Effect of pH and temperature on the binding of bilirubin to human erythrocyte membranes.

Effect of pH and temperature on the binding of bilirubin to human erythrocyte membranes was studied by incubating the membranes at different pH and temperatures and determining the bound bilirubin. At all pH values, the amount of membrane-bound bilirubin increased with the increase in bilirubin-to-albumin molar ratios (B/As), being highest at lower pH values in all cases. Further, linear increase in bound bilirubin with the increase in bilirubin concentration in the incubate was observed at a constant B/A and at all pH values. However, the slope value increased with the decrease in pH suggesting more bilirubin binding to membranes at lower pH values. Increase in bilirubin binding at lower pH can be explained on the basis of increased free bilirubin concentration as well as more conversion of bilirubin dianion to monoanion. Temperature dependence of bilirubin binding to membranes was observed within the temperature range of 7 degrees -60 degrees C, showing minimum binding at 27 degrees C and 37 degrees C which increased on either side. Increase in bilirubin binding at temperatures lower than 20 degrees C and higher than 40 degrees C can be ascribed to the change in membrane topography as well as bilirubin-albumin interaction.

Bilirubin↗
PubMed1999

A comparative study on the extraction of membrane-bound bilirubin from erythrocyte membranes using various methods.

In this study, we used three different methods for the extraction of membrane-bound bilirubin (EMB) from erythrocyte membranes. Use of 2.5% albumin, pH 7.4, for elution of EMB resulted in only 34% of the total EMB which was estimated after the solubilization of bilirubin-loaded erythrocyte membranes (BLEMs) with 1% SDS. On the other hand, incubation of BLEMs with 38 mM sodium carbonate solution containing 5 mM EDTA, pH 11.0, yielded 77% of the total EMB. Application of Fog's reaction for the estimation of EMB directly on the BLEMs resulted in the estimation of 75% of the total EMB. These results suggest that either of the above methods, i.e. use of albumin or high pH, or direct Fog's reaction cannot estimate the total EMB correctly. Increase in ionic strength from 0.15 to 0.45 did not release any EMB from erythrocyte membranes. Therefore, the best method for the estimation of total EMB is the solubilization of membrane with 1% SDS followed by Fog's reaction method.

Bilirubin↗
PubMed1999

Doxorubicin-induced congestive heart failure in elderly patients with metastatic breast cancer, with long-term follow-up: the M.D. Anderson experience.

PURPOSE: Correlation between aging and doxorubicin-induced congestive heart failure in patients with metastatic breast cancer was studied to determine whether doxorubicin-induced congestive heart failure in elderly patients with metastatic breast cancer is a clinically significant issue. METHODS: This was a retrospective study with a median follow-up of 16.8 years. The setting was a comprehensive cancer center in a large city. A group of 682 consecutive patients with metastatic breast cancer presented to The University of Texas M.D. Anderson Cancer Center between 1973 and 1980. All patients received doxorubicin by bolus infusion. Patients in group 1 (n = 538) were aged 50 to 64 years; patients in group 2 (n = 144) were aged 65 years and older. Medical records of all patients were reviewed. Patients who had congestive heart failure were identified and analyzed. The diagnosis of doxorubicin-induced congestive heart failure was made and confirmed by a cardiologist at the time of its development. The main outcome measure was the cumulative probability of developing doxorubicin-induced congestive heart failure in elderly patients with metastatic breast cancer compared to a younger age group. RESULTS: In group 1, 33 patients, and in group 2, 13 patients developed doxorubicin-related congestive heart failure. The cumulative doses of doxorubicin administered to patients with congestive heart failure were 410 mg/m2 (range 150-550 mg/m2) and 400 (range 100-570 mg/m2), respectively. The time interval from the last date of doxorubicin treatment to the development of congestive heart failure was, respectively, 5 months (range < 1-65 months) and 9 months (range < 1-28 months). There was no statistically significant difference between the two congestive heart failure subgroups, nor were we able to identify risk factors that could have increased the risk of congestive heart failure among these patients. CONCLUSION: Older patients with metastatic breast cancer and no significant comorbidity can be treated with doxorubicin-based chemotherapy with no added risk of developing congestive heart failure beyond that in the younger age group.

Aged↗
PubMed1999

Differential resistance to calcium-induced bilirubin-dependent hemolysis in mammalian erythrocytes.

Washed erythrocytes from human, buffalo, sheep and goat preincubated with different concentrations of calcium chloride (16.7-1830 microM) showed significantly different rates of hemolysis (up to 62%) after addition of bilirubin (72 microM). Goat erythrocytes displayed marked resistance to hemolysis with only 11% hemolysis observed at the highest calcium concentration. Similar trend in hemolysis was also observed when the concentration of CaCl2 was fixed (330 microM) and bilirubin concentration varied (0-72 microM). (Ca(2+)-Mg(2+)-ATPase levels were found significantly lower in goat and sheep erythrocyte membranes compared to human and buffalo erythrocyte membranes. This was correlated well with the observed hemolysis in various mammalian erythrocytes.

Adenosine Triphosphate↗