Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Linear Regression”

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 613 records · Page 34Linked to original sources
PubMed2006

[Normality tables in the Spanish population for psychometric tests used in the diagnosis of minimal hepatic encephalopathy].

BACKGROUND AND OBJECTIVE: Minimal hepatic encephalopathy (MHE) induces an impairment in the quality of life and has been found as the first stage in hepatic encephalopathy syndrome. Psychometric Hepatic Encephalopathy Score (PHES) is a psychometric test battery widely utilized in the diagnosis of MHE which requires correction by age and education. POPULATION AND METHOD: Eight-hundred and eighty four healthy people were included. Trailmaking test-A (TMT-A), TMT-B, digit symbol test (DST), serial dotting (SDT) and line drawing test (LDT) were performed. Age, education, gender, alcohol consumption and hospitalization were recorded. T-student and Pearson correlation were utilized in univariate analyses. A multiple linear regression was performed to detect independent variables associated with psychometric test score. Using the equation from the multiple linear regression, normality tables were built. RESULTS: Age and education were associated with all 5 tests. An alcohol consumption higher than 10 g per day was associated with a lower score in DST and SDT. A gender influence was seen in TMT-A and TMT-B. Moreover, some differences were found in SDT, TMT-B and LDT according to hospital procedence. In multivariate analyses using multiple linear regression, age and education were independently associated with all five tests. CONCLUSIONS: Normality tables of the PHES allow investigating MHE in cirrhotics avoiding local control groups. Taking in mind normality tables, PHES could become a gold standard test to analyze further diagnostic methods and to evaluate drugs efficacy in MHE. Normality tables are freely available at http://www.redEH.org.

Adult↗
PubMed1999

Plasmin-alpha2-antiplasmin complex in patients with atrial fibrillation. Stroke Prevention in Atrial Fibrillation Investigators.

Plasmin-alpha2-antiplasmin complex (PAP) is an index of recent fibrinolytic activity. We examined PAP levels in patients with atrial fibrillation (AF) to determine whether these levels are correlated with clinical characteristics associated with stroke risk. We obtained blood for measurement of PAP in a non-random sample of 586 patients with AF on entering the Stroke Prevention in Atrial Fibrillation III Study. PAP levels were measured with an ELISA assay. PAP values were transformed with a natural logarithm (PAPln) prior to all analyses. Older age, female gender, recent congestive heart failure, decreasing fractional shortening, recent onset of AF, and coronary artery disease were each univariately associated with higher levels of PAP (all p<0.05, two-sample t-test, simple linear regression). Older age, recent congestive heart failure, decreasing fractional shortening, and recent onset of AF were independently associated with higher PAP levels by multivariate analysis (linear regression). Among patients receiving warfarin, PAP levels were not correlated with INR levels (linear regression, p=0.60). Patients classified as high-risk for thromboembolism by our risk stratification criteria (systolic blood pressure > 160 mm Hg, prior thromboembolism, recent congestive heart failure, poor left ventricular function, and women over age 75) had higher PAP levels than low-risk patients (antilog mean PAPln 5.6 vs 4.9. p<0.001, two-sample t-test). PAP levels in patients with AF are associated with clinical characteristics predictive of thromboembolism. Elevated PAP levels are particularly associated with poor left ventricular function and are not affected by anticoagulation. PAP levels may be a marker of stroke risk in patients with AF.

Aged↗
PubMed2006

Multislice CT angiography in aortic stent grafting: relationship between image noise and body mass index.

PURPOSE: To investigate the correlation between image noise and body mass index (BMI) in multislice CT angiography (MSCT) for patients with abdominal aortic aneurysm (AAA) treated with endovascular stent grafts. MATERIALS AND METHODS: Seventeen patients who underwent MSCT following endovascular repair of AAA were included in the study. Image noise (standard deviation of the CT attenuation: S.D.) and signal to noise ratio (SNR) were plotted against BMI to demonstrate the correlation using a linear regression method. Image quality of 3D reconstructions was correlated to the SNR and BMI. RESULTS: The r-value of linear regression between S.D. and BMI was 0.578 (p<0.05), 0.835 and 0.802 (p<0.001), respectively, at the level of renal artery, aortic aneurysm and common iliac artery. The r-value of linear regression between SNR and BMI was 0.332, 0.516 and 0.552 (p<0.05), respectively, at above three levels. Image quality of 3D reconstructions was compromised in five patients and diagnosis was affected in two patients with BMI more than 30. CONCLUSION: A significant correlation was observed between image noise and BMI in MSCT angiography of endovascular repair of AAA. Our findings are valuable for optimisation of MSCT angiography scanning protocols and reduction of radiation dose in MSCT examinations.

Aged↗
PubMed1990

A discriminatory study of a pharmacokinetic model for intramuscular gentamicin in sheep.

The resulting serum concentrations were measured in six ewes after intramuscular administration of 10 mg/kg of gentamicin. The model providing the best fit for the experimental data was determined both by linear regression analysis between the experimental and theoretical values and by means of the Minimum Akaike Information Criterion Estimation (MAICE) test. Linear regression analysis showed certain differences favouring the monocompartmental model although the advantage was not conclusive. The MAICE test, however, permitted a clear discrimination in favour of the same model. When linear regression analysis is not conclusive, the MAICE test represents a good alternative.

Animals↗
PubMed2006

Pushing the boundaries of viability: the economic impact of extreme preterm birth.

BACKGROUND: Previous assessments of the economic impact of preterm birth focussed on short term health service costs across the broad spectrum of prematurity. OBJECTIVE: To estimate the societal costs of extreme preterm birth during the sixth year after birth. METHODS: Unit costs were applied to estimates of health, social and broader resource use made by 241 children born at 20 through 25 completed weeks of gestation in the United Kingdom and Republic of Ireland and a comparison group of 160 children born at full term. Societal costs per child during the sixth year after birth were estimated and subjected to a rigorous sensitivity analysis. The effects of gestational age at birth on annual societal costs were analysed, first in a simple linear regression and then in a multiple linear regression. RESULTS: Mean societal costs over the 12 month period were 9541 pounds sterling (standard deviation 11,678 pounds sterling) for the extreme preterm group and 3883 pounds sterling (1098 pounds sterling) for the term group, generating a mean cost difference of 5658 pounds sterling (bootstrap 95% confidence interval: 4203 pounds sterling, 7256 pounds sterling) that was statistically significant (P<0.001). After adjustment for clinical and sociodemographic covariates, sex-specific extreme preterm birth was a strong predictor of high societal costs. CONCLUSION: The results of this study should facilitate the effective planning of services and may be used to inform the development of future economic evaluations of interventions aimed at preventing extreme preterm birth or alleviating its effects.

Cohort Studies↗
PubMed2004

Evaluation of internal lung motion for respiratory-gated radiotherapy using MRI: Part II-margin reduction of internal target volume.

PURPOSE: To analyze the relationship between lung motion and skin surface motion during respiration, determine the uncertainties and variability of such a relationship, and assess the potential of reducing internal target margin for gated radiotherapy. METHODS AND MATERIALS: Three healthy volunteers and four lung cancer patients were recruited in a prospective imaging study using MRI to track the internal lung and external skin motion during breathing. The relationship between the lung and skin motion was modeled using linear regression analysis. The slope of the linear fit and its confidence interval were analyzed for different lung locations, skin surface locations, and breathing patterns from separate imaging sessions. The margins of the internal target volume were calculated based on the residual lung motion during gating and its uncertainties from multiple treatment fractions for the gated treatment. RESULTS: The slope and confidence interval of the linear regression from the motion analysis were uniquely defined by the locations of the lung, skin surface, and breathing patterns. Statistically significant differences were observed among individuals and between different times of measurement. The normal free-breathing motion averaged from all volunteer and patient data was 13.4 +/- 7.4 mm along the superior-inferior (SI) direction and 6.9 +/- 2.6 mm along the anterior-posterior (AP) direction. With simulated respiratory gating, the average margin reduction was 5.5 +/- 4.8 mm and 1.6 +/- 1.0 mm, respectively, along the SI and AP directions (or 36% +/- 15% and 25% +/- 14%, respectively, relative to free-breathing motion). CONCLUSION: Because respiratory movement is rather complex, the relationship between the lung and skin surface motion is affected by many anatomic and physiologic factors. The reduction of internal target margin and efficacy of the free-breathing gating technique should be assessed for individual cases.

Adult↗
PubMed2005

Racial disparities in the association of foetal growth retardation to childhood blood pressure.

BACKGROUND: Foetal growth retardation (FGR), defined as less than the 10th percentile of birth weight for gestational age, is reported to be an important contributor to hypertension and cardiovascular disease in children and adults, but findings are not consistent. For this reason we re-examined the role of FGR in childhood blood pressure. METHODS: We performed univariate and multivariate analyses on data gathered from 262 children, age 5 years, born to mothers at risk for pre-term delivery or FGR infant. The characteristics of the mothers and the children were evaluated using Student's t-test. Rates and proportions were compared using either chi-square or Fisher's exact test. Linear regression models evaluated the effect of birth weight and body mass index on systolic and diastolic blood pressure. Multivariate linear regression was used to model the effects of FGR, gestational age, body mass index, race, gender, maternal smoking, maternal gestational diabetes on blood pressure while adjusting for possible confounders. RESULTS: Systolic blood pressure was inversely associated with birth weight in white children while a small direct association was noted in African Americans. Body mass index was positively associated with systolic blood pressure in both groups. Multiple linear regression analyses showed FGR and early gestational age were associated with higher blood pressure in white but not African American children, accounting for a 13.2 mmHg difference between FGR and appropriate for gestational age groups. Blood pressure in African Americans was strongly affected by maternal gestational diabetes and smoking. CONCLUSIONS: Birth weight influences childhood blood pressure but the effects may vary depending on ethnic group. The relative importance of birth weight on blood pressure may depend on other prenatal and post-partum risks.

Adult↗
PubMed2005

Preoperative prostate specific antigen doubling time and velocity are strong and independent predictors of outcomes following radical prostatectomy.

PURPOSE: Prostate specific antigen (PSA) is a useful marker for predicting outcomes following treatment for prostate cancer but, given the evolving nature of prostate cancer, there is an ongoing need to refine its use. We assessed preoperative PSA doubling time (PSADT) and PSA velocity (PSAV) as predictors of outcome following radical retropubic prostatectomy (RRP). MATERIALS AND METHODS: We identified 2,290 men who were treated with RRP for prostate cancer between 1990 and 1999 with multiple preoperative PSA measurements available. PSADT was calculated by log linear regression and PSAV was calculated by linear regression. These parameters were used in preoperative and postoperative multivariate models for the end points of biochemical and clinical progression, and cancer death. RESULTS: At a median followup of 7.1 years (range 0.1 to 14.5) biochemical progression, clinical progression and death from prostate cancer were observed in 583, 156 and 42 patients, respectively. The HR for death from prostate cancer was 6.22 (95% CI 3.33 to 11.61) in men with PSADT less than 18 months vs 18 or greater and 6.54 (95% CI 3.51 to 12.19) in men with PSAV greater than 3.4 ng/ml yearly vs 3.4 or less. On multivariate analysis adjusting for preoperative or postoperative variables PSADT and PSAV remained significant predictors of each outcome. When assessed jointly, PSAV was significant as a predictor of biochemical progression, while PSADT was a significant predictor of clinical progression and cancer death. CONCLUSIONS: This study confirms the usefulness of preoperative PSA kinetics for predicting post-RRP outcomes, which may be useful for stratifying patients, so that rational management decisions can be made with respect to observation, intervention and adjuvant treatment. While PSADT maybe biologically more accurate and stronger on multivariate analysis, PSAV is clinically easier to use and a good approximation in the short term.

Adult↗
PubMed2006

Postchallenge glucose rises with increasing age even when glucose tolerance is normal.

AIMS: Ageing increases the likelihood of developing diabetes, with associated cardiovascular disease. In a cross-sectional study, we sought to determine whether age is associated with an increase in glucose concentrations 1 h after an oral glucose challenge (1-h OGTT), even when glucose tolerance is normal (NGT). METHODS: Among subjects in the NHANES II database, 2591 subjects with NGT and documented 1-h OGTT glucose concentrations were studied. The relationship between age and 1-h OGTT glucose concentrations was assessed in a multivariable linear regression analysis. RESULTS: In a multivariable linear regression analysis, each 10-year increase in age conferred an additional 0.20 mmol/l increase in the 1-h OGTT glucose (P < 0.0001). Moreover, an interaction between age and gender was found such that 1-h OGTT glucose concentrations rose more rapidly with increasing age in men than in women. The impact of age on 1-h OGTT glucose was independent of both fasting and 2-h OGTT glucose concentrations. CONCLUSIONS: One-hour OGTT glucose concentrations rise significantly with age even in subjects with NGT. Further investigation is warranted to explore the pathophysiological significance of such age-related impairment of glucose handling, which might increase the risk of cardiovascular disease even when patients do not meet criteria for the diagnosis of diabetes or prediabetes.

Adolescent↗
PubMed1981

Correlation of moxalactam (LY127935) susceptibility tests by disk diffusion and agar plate dilution methods.

Susceptibility of pathogenic aerobic bacteria to moxalactam (LY127935) was compared by two methods, diffusion from 30-microgram disks and agar plate dilution. The two methods gave a satisfactory degree of correlation when compared by linear regression, but the slope of the linear regression was significantly steeper for gram-negative bacilli (Enterobacteriaceae and Pseudomonas aeruginosa) than for coccal organisms (streptococci and staphylococci). The largest zone diameters by disk diffusion were found with Citrobacter, Proteus, Escherichia coli, Klebsiella, and Enterobacter strains. The error rate-bounded method of Metzler and DeHaan (J. Infect. Dis. 130:588-594, 1974) gave useful breakpoints of disk zone sizes for moxalactam resistance and susceptibility.

Anti-Bacterial Agents↗
PubMed1988

Productive longevity of first-cross cows produced in a five-breed diallel: I. Reasons for removal.

Data from 498 cows of 15 breed-types produced in a five-breed diallel (reciprocals pooled) including Angus, Brahman, Hereford, Holstein and Jersey were analyzed for differences in productive longevity, mean life span and reasons for removal from the herd over approximately 14 yr. The only culling related to production was of cows that did not give birth to at least one live calf every 24 mo. Otherwise, cows were only removed when they appeared to be incapable of weaning another live calf. The average age at removal, estimated from linear regression, was 3,470 d, with values ranging from 2,174 d (Jersey) to 4,087 d (Angus-Brahman). Longevity of crossbred cows was greater than that of purebred cows (P less than .001). Purebred cows usually had higher initial rates of removal than crossbred cows. Estimates of mean longevity from the nonlinear regression resembled the results of the linear regression analysis; mean life spans ranged from 6.2 yr (Jersey) to 14.6 yr (Angus-Brahman). The major reason for removal of Brahman and Brahman-cross cows was for reproductive reasons. Angus and Angus-cross cows were culled more frequently for structural unsoundness; Hereford cows had a higher incidence of cancer eye, calving difficulty and mammary problems; cows with 100% dairy breeding had exceptionally high susceptibility to diseases and mineral imbalance.

Animals↗
PubMed1990

[Regression and correlation].

Simple linear regression and correlation are frequently used in medical publications. The methods are overused and have become greatly confused, probably because of the close similarity between the mathematical calculations. Simple regression is a method for a mathematical description of the dependence of a random response variable of another variable, a regressor, which is not random. The model can be used to predict the response form given values of the regressor. Correlation is a measure of the degree of linear association between two random variables, and has nothing to do with regression. The method is mainly an investigative tool in a statistical analysis, or to suggest further research; for forming hypotheses rather than for testing them. To judge form publications, fitting of a regression line and calculation of the correlation coefficient is the method of choice in modelling the linear relationship between two random variables. This is a completely misguided approach. When both variables are random, least squares estimation, as used in regression, is not appropriate. The correct method is to estimate a linear structural model by maximum likelihood.

Data Interpretation, Statistical↗
PubMed1986

Quantification of analyte and interferent by multipoint analysis.

We have investigated the application of multipoint kinetic curve-fitting methods to the determination of an analyte in the presence of a single interferent. Our model system for the analyte-interferent was creatinine-acetoacetate as determined with the kinetic Jaffé method. We examined the utility of the following multipoint approaches: simultaneous equations, multivariable linear regression, and iterative multivariable nonlinear regression. With appropriate restrictions, all approaches could detect acetoacetate interference and quantify both creatinine and acetoacetate. A two-stage linear regression approach was both versatile and computationally simple. Interferent was detected in the first stage, and both analyte and interferent were quantified in the second stage if the interferent was assumed known and an adequate fit of the model to the data was obtained. Using the two-stage linear regression model, we obtained results for 10 ketotic patients that correlated well with results by enzymatic methods for creatinine (r = 0.976) and acetoacetate (r = 0.995); we also demonstrated that creatinine could be quantified in the presence of the antibiotic cefoxitin.

Acetoacetates↗
PubMedDate not supplied

Age and sex dependency of the Achilles tendon reflex.

Achilles tendon reflex time was measured in normal subjects (1,837 females and 7,937 males). Values of half relaxation time showed a clear dependency on both age and sex. The reflex becomes slower with advancing age and was consistently slower in females. The results redefine the normal range of the Achilles tendon reflex. Regression lines of relaxation time on age were fitted for each sex separately. Lines representing linear regression, log-linear regression or quadratic regression were inferior to those described by linear splines. The gradual increase of Achilles tendon relaxation time with age was not due to unrecognized hypothyroidism or peripheral neuropathy. It is suggested that one cause of the progressive increment is diminution of sympathetic tone with age, and changes in the muscles themselves may also contribute.

Achilles Tendon↗
PubMed1999

Serum leptin levels in healthy ageing men: are decreased serum testosterone and increased adiposity in elderly men the consequence of leptin deficiency?

OBJECTIVE: The limited information on serum leptin levels in elderly men suggests the occurrence of an age-related decrease, with disruption of the relationship between fat mass and leptin levels. A relative leptin deficiency might thus be implicated in the increase of fat mass and decrease of serum testosterone levels in elderly men. Therefore, we have reevaluated the age-related changes in serum leptin levels and their relationship with adiposity and androgen levels in a large group of community dwelling men. SUBJECTS AND MEASUREMENTS: Serum leptin and androgen levels were measured in 271 healthy, ambulatory elderly men (median age 74 years), as well as in 61 middle-aged (median 43 years) and 40 young (median 25.5 years) controls. Adiposity was assessed by anthropometrical measurements (body mass index; BMI) and by estimation of fat mass by the bio-impedance method. RESULTS: Serum leptin levels, whether or not adjusted for BMI, were found to increase with age, the values tending to level off after the age of 45 years, and were strongly correlated to BMI (r = 0.77) and fat mass assessed by the bio-impedance method (r = 0.81). Linear regression analysis showed a similar slope for the relationship between BMI and serum leptin in the three age groups. Multiple linear regression analysis indicated BMI, age and serum insulin, but not serum testosterone, as significant independent correlates of serum leptin. Serum (free) testosterone levels were negatively correlated with age and serum leptin, also after partialization for BMI: rank correlation coefficients vs. age and serum leptin, respectively, were - 0.20 (P < 0.001) and - 0.16 (P < 0.01) for total testosterone and - 0.60 (P < 0.001) and - 0.23 (P < 0.001) for free testosterone. Dehydroepiandrosterone sulphate (DHEAS) and leptin levels emerged as significant independent correlates in a multiple linear regression model for total serum testosterone; BMI and serum insulin became highly significant correlates in the same model when leptin was omitted from the independent variables. CONCLUSION: Ageing in men is accompanied by a rise of serum leptin levels with a maintained strong association between serum leptin and adiposity in elderly men. Testosterone does not appear to be a major determinant of serum leptin in healthy men, while leptin does emerge as a negative correlate of serum testosterone. Increased fat mass and decreased testosterone production in elderly men cannot be attributed to a relative leptin deficiency.

Adult↗
PubMed2003

Comparison of myocardial fatty acid metabolism with left ventricular function and perfusion in cardiomyopathies: by 123I-BMIPP SPECT and 99mTc-tetrofosmin electrocardiographically gated SPECT.

OBJECTIVE: To investigate myocardial fatty acid metabolism and its relationship with left ventricular (LV) function and perfusion in hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy (DCM). METHODS: Thirty-nine patients with cardiomyopathies (58 +/- 14 y), comprising 15 DCM and 24 HCM, and 9 age-matched healthy controls were studied with 123I-15-(p-iodophenyl)-3-(R,S)-methylpentadecanoic acid (BMIPP) and 99mTc-tetrofosmin (TF) electrocardiographically gated SPECT. As parameters of myocardial fatty acid metabolism, the heart-to-mediastinum ratio (H/M) and global washout of BMIPP were calculated from early and delayed planar images, while regional BMIPP uptake and washout were calculated from SPECT. In TF study, the H/M (H/M-TF) and LV ejection fraction (LVEF) were calculated as global parameters of perfusion and function, while regional TF uptake and wall thickening index were calculated as regional parameters of perfusion and function using the Quantitative Gated SPECT software. The differences in the parameters and the correlations between the parameters from the 2 studies were investigated by one-way ANOVA and multiple linear regression analysis. RESULTS: BMIPP uptake was decreased (p < 0.05), and its washout was increased (p < 0.05) in DCM and HCM. In multiple linear regression analysis, global BMIPP parameters showed no significant correlation with LVEF (p > 0.05), but showed a significant correlation with H/M-TF (p < 0.05) in DCM and HCM. According to the partial correlation coefficient, early H/M was the only significant factor (p < 0.05) for predicting H/M-TF in DCM and HCM. Multiple linear regression analysis on regional parameters showed regional BMIPP parameters had no correlation with regional function (p > 0.05) but had a significant correlation with regional perfusion (p < 0.0001) in DCM. In HCM, regional BMIPP parameters showed significant multiple linear correlations with both regional function (p < 0.005) and perfusion (p < 0.0001). According to the partial correlation coefficients, delayed regional BMIPP uptake was the most significant factor for predicting regional function in HCM, while early regional BMIPP uptake was the only or the most significant factor for predicting regional perfusion in DCM and HCM, respectively. CONCLUSION: In DCM, BMIPP uptake and washout could not reflect LV function. In HCM, regional delayed BMIPP uptake might be useful for evaluating regional function. In DCM and HCM, early BMIPP uptake might be largely determined by myocardial perfusion.

Adolescent↗
PubMed2006

Time of death dependent criteria in vitreous humor: accuracy of estimating the time since death.

Since more than 40 years reports on the rise of potassium concentration [K(+)] in vitreous humor have been published with different statements concerning the accuracy of death time estimation. In the last years several statistical approaches for a more accurate estimation of the time since death have been reported. While in most investigations the PMI has been used as the independent and [K(+)] as the dependent variable in linear regression analysis between PMI and [K(+)], recently it has been suggested to use [K(+)] as the independent variable for regression analysis. Changing the variables would lead to a higher accuracy of death time estimation. This has also been recommended for regression analysis between hypoxanthine concentration [Hx] and time since death. This hypothesis has been checked on independent cases with potassium and hypoxanthine in vitreous humor. Linear regression with [K(+)] or [Hx] as independent variable has revealed a slightly more accurate death time estimation compared to a linear regression with PMI as independent variable. Thus, e.g. the accuracy could have been improved from +/-25.96 to +/- 23.27h by using [K(+)] as independent variable. Another statistical approach has re-evaluated six large studies on the rise of vitreous [K(+)] using a local regression analysis (Loess procedure). Based on this re-evaluation an accuracy of death time estimation has been recommended (95% limits of confidence of +/-1h in the early PMI and +/-10h, 110h postmortem) which has surpassed even optimistic results of earlier investigations. This recommended accuracy of death time estimation has been checked on a random sample of 492 cases. Only 153 cases have been within the predicted postmortem interval, 339 lay outside with a systematic overestimation of the time since death.

Chromatography, High Pressure Liquid↗
PubMed2003

Limitations in linearized analyses of binding equilibria: binding of TNP-ATP to the H4-H5 loop of Na/K-ATPase.

Binding of TNP-ATP [2',3'- O-(2,4,6-trinitrophenyl)adenosine 5'-triphosphate, a fluorescent analogue of ATP] to the K605 protein was studied. This is an isolated N-domain in the cytoplasmic loop of the Na/K-ATPase alpha-subunit, lying between membrane-spanning segments 4 and 5 (sequence L(354)-I(604)). A titration equation is derived that explicitly makes it possible to relate the fluorescence of TNP-ATP and K605 solutions to total probe concentration in the sample. Using this, it is possible to obtain the value of the dissociation constant from the titration experiment without resorting to the Scatchard plot, which is far from optimal from the statistical point of view. Using the new formula with non-linear regression analysis, a value of the dissociation constant K(D) for TNP-ATP binding to the K605 protein of 3.03 +/- 0.28 microM at 22 degrees C was obtained. A series of fits to simulated data with added noise demonstrated clearly the advantage of non-linear regression using the new formula over the commonly employed linear regression using the Scatchard plot. The procedure presented is generally applicable to binding assays using changes in the fluorescence of ligands on binding.

Adenosine Triphosphate↗