Cross-product Terms Research Articles

Dynamic analysis, circuit implementation and passive control of a novel four-dimensional chaotic system with multiscroll attractor and multiple coexisting attractors

In this paper, we construct a novel 4D autonomous chaotic system with four cross-product nonlinear terms and five equilibria. The multiple coexisting attractors and the multiscroll attractor of the system are numerically investigated. Research results show that the system has various types of multiple attractors, including three strange attractors with a limit cycle, three limit cycles, two strange attractors with a pair of limit cycles, two coexisting strange attractors. By using the passive control theory, a controller is designed for controlling the chaos of the system. Both analytical and numerical studies verify that the designed controller can suppress chaotic motion and stabilise the system at the origin. Moreover, an electronic circuit is presented for implementing the chaotic system.

Acoustic nonreciprocity in Coriolis mean flow systems.

One way to break acoustic reciprocity is to have a moving wave propagation medium. If the acoustic wave vector and the moving fluid velocity are collinear, the wave vector shift caused by the fluid flow can be used to break. In this paper, an alternative approach is investigated in which the fluid velocity enters the differential equation of the system as a cross product term with the wave vector. A circular field where the fluid velocity increases radially has a Coriolis acceleration term. In such a system, the acoustic wave enters from the central wall and exits from the perimeter wall. In this paper, the differential equation is solved numerically and the effect of fluid velocity on the nonreciprocity factor is examined.

Associations of Four Community Factors With Longitudinal Change in Hemoglobin A1c Levels in Patients With Type 2 Diabetes.

OBJECTIVETo evaluate associations of community factors with glycated hemoglobin (HbA1c).RESEARCH DESIGN AND METHODSWe identified patients with type 2 diabetes who had an HbA1c ≥7.5% (58 mmol/mol) and subsequent HbA1c testing within 90–270 days. We used mixed-effect models to assess whether treatment intensification (TI) and community domains (community socioeconomic deprivation [CSD], food availability, fitness assets, and utilitarian physical activity favorability [quartiled]) were associated with HbA1c change over 6 and 24 months, controlling for demographics, HbA1c, BMI, and time with evidence of type 2 diabetes. We evaluated whether community domains modified associations of TI with HbA1c change using cross product terms.RESULTSThere were 15,308 patients with 69,818 elevated HbA1c measures. The average reduction in HbA1c over 6 months was 0.07% less in townships with a high level of CSD (third quartile versus the first). Reductions were 0.10% greater for HbA1c in townships with the best food availability (versus worst). HbA1c reductions were 0.17–0.19% greater in census tracts in the second and third quartiles of utilitarian physical activity favorability versus the first. The association of TI with 6-month HbA1c change was weaker in townships and boroughs with the worst CSD (versus best) and in boroughs with the best fitness assets (versus worst). The association of TI with 24-month HbA1c change was weaker in census tracts with the worst CSD (versus third quartile) and strongest in census tracts most favorable for utilitarian physical activity (versus worst).CONCLUSIONSCommunity domains were associated with HbA1c change and blunted TI effectiveness.

The association between neighborhood socioeconomic disadvantage and high-risk injection behavior among people who inject drugs

BackgroundAlthough much research has been conducted on the determinants of HIV risk behavior among people who inject drugs (PWID), the influence of the neighborhood context on high-risk injection behavior remains understudied. To address this gap in the literature, we measured associations between neighborhood socioeconomic disadvantage and high-risk injection behavior, and determined whether these associations were modified by drug-related police activity and syringe exchange program (SEP) accessibility. MethodsOur sample was comprised of 484 pharmacy-recruited PWID in New York City. Measures of neighborhood socioeconomic disadvantage were created using data from the 2006–2010 American Community Survey. Associations with high-risk injection behavior were estimated using multivariable Poisson regression. Effect modification by drug-related police activity and SEP accessibility was assessed by entering cross-product terms into adjusted models of high-risk injection behavior. ResultsNeighborhood socioeconomic disadvantage was associated with decreased receptive syringe sharing and unsterile syringe use. In neighborhoods with high drug-related police activity, associations between neighborhood disadvantage and unsterile syringe use were attenuated to the null. In neighborhoods with high SEP accessibility, neighborhood disadvantage was associated with decreased acquisition of syringes from an unsafe source. ConclusionsPWID in disadvantaged neighborhoods reported safer injection behaviors than their counterparts in neighborhoods that were relatively better off. The contrasting patterns of effect modification by SEP accessibility and drug-related police activity support the use of harm reduction approaches over law enforcement-based strategies for the control of blood borne virus transmission among PWID in disadvantaged urban areas.

Joint Effects of PON1 Polymorphisms and Vegetable Intake on Ischemic Stroke: A Family-Based Case Control Study.

Paraoxonase 1 gene (PON1) polymorphisms and dietary vegetable and fruit intake are both established determinants of ischemic stroke (IS). However, little is known about whether these factors jointly influence the risk of IS. We analyzed the main effects of PON1, as well as the interactions between PON1 and dietary vegetable or fruit intake with the risk of total IS and its subtypes in a family-based case-control study conducted among 2158 Chinese participants (1007 IS cases and 1151 IS-free controls) from 918 families. Conditional logistic regression models, with each family as a stratum, were used to examine the association between rs662 and IS. Gene-diet interactions were tested by including a cross-product term of dietary vegetable or fruit intake by rs662_G allele count in the models. Each copy of the PON1 rs662_G allele was associated with 28% higher risk of total IS (p = 0.008) and 32% higher risk of large artery atherosclerosis subtype (LAA) (p = 0.01). We observed an interaction between rs662 and vegetable intake for both total IS (p = 0.006) and LAA (p = 0.02) after adjustment for covariates. Individuals who carry the rs662_A allele may benefit to a greater extent from intake of vegetables and thus be more effectively protected from ischemic stroke, whereas carriers of the G allele may still remain at greater risk for ischemic stroke due to their genetic backgrounds even when they consume a high level of vegetables. More studies are needed to replicate our findings among other populations.

Risk factors for breast cancer in the breast cancer risk model study of Guam and Saipan

BackgroundChamorro Pacific Islanders in the Mariana Islands have breast cancer incidence rates similar to, but mortality rates higher than, those of U.S. women. As breast cancer risk factors of women of the Mariana Islands may be unique because of ethnic and cultural differences, we studied established and suspected risk factors for breast cancer in this unstudied population. MethodsFrom 2010–2013, we conducted retrospective case-control study of female breast cancer (104 cases and 185 controls) among women in the Mariana Islands. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each of various lifestyle-related factors from logistic regression of breast cancer, in all women and in pre- and postmenopausal women separately. Tests for interaction of risk factors with ethnicity were based on the Wald statistics for cross-product terms. ResultsOf the medical and reproductive factors considered — age at menarche, breastfeeding, number of live births, age at first live birth, hormone use, and menopause — only age at first live birth was confirmed. Age at first live birth, among parous women, was higher among cases (mean 24.9 years) than controls (mean 23.2 years); with increased breast cancer risk (OR=2.53; 95% CI, 1.04–6.19 for age≥30y compared to <20y, P for trend=0.01). Of the lifestyle factors —body mass index, waist circumference, physical activity, alcohol and betel-nut intake, and education — only waist circumference (OR=1.65; 95% CI 0.87–3.14 for the highest tertile group compared to the lowest, P for trend=0.04) was significantly associated with breast cancer risk and only in Filipino women. The association with many other established risk factors, such as BMI, hormone use and physical activity, were in the expected direction but were not significant. Associations for family history of breast cancer and alcohol intake were not evident ConclusionsThe results provide a basis for cancer prevention guidance for women in the Mariana Islands.

Optimization of protein extraction from tree peony (Paeonia suffruticosa Andr.) seed using response surface methodology

Optimization of extraction ratio (ER) of tree peony seed protein (TPSP) was investigated using response surface methodology (RSM). The second-degree equation for ER of TPSP had high coeffi cient (0.9625) of determination. The probability (P) value of regression model signifi cance was less than 0.001 by analysis of central composite rotatable design. Relationships of ER to pH, liquid/solid ratio, squares of all factors, and cross-product factors (x2x3, x2x4, x3x4) were signifi cant (P&lt;0.05). Whereas, extraction time, temperature, and cross-product terms (x1x2, x1x3, x1x4) were not signifi cant factors (P&gt;0.05). Optimum extraction conditions were 3.42 h, pH 9.50, 50.80 ºC, and 9.54 ml g–1 of liquid/solid ratio with the maximum ER (43.60%) . SDS-PAGE indicated TPSP had mainly four proteins (180, 100, 60, and 35 kDa) with four subunits of 60, 48, 38, and 23 kDa. TPSP had a good amino acid composition with abundant essential amino acids (39.76%) determined by amino acid analysis.

IFRS convergence and revisions: value relevance of accounting information from East Africa

PurposeThe purpose of this paper is to examine the relative value relevance of accounting information arising from the adoption of converged and revised International Accounting Standards (IAS)/International Financial Reporting Standards (IFRS) in East Africa.Design/methodology/approachThe research applies “same firm year” design for identification of the effects of changes in accounting standards. A model similar to Ohlson’s price model and random-effects GLS are used to estimate R2 of the regressions of share prices on book values and earnings.FindingsThe results show that accounting information prepared from revised and converged IAS/IFRS display higher value relevance and also increased following the revision and convergence of IAS/IFRS. The cross-product term is more significant in the post-revision/convergence period thus providing further evidence for increased value relevance after the revision of IAS/IFRS. The results are robust to various models and show that value relevance in East Africa is relatively lower than that of the developed markets.Originality/valueThe current study provides empirical evidence that value relevance increases with converged/revised IAS/IFRS based on quasi natural experimental setting in East Africa. The authors also extend the debate on whether value relevance is relevant in emerging markets, which are regarded as imperfect markets with few regulations, weak enforcement and limited sources of information. The results may be useful to accounting preparers, regulators, investors, standard setters and countries seeking to adopt IAS/IFRS in developing countries.

Amplitude Control Analysis of a Four-Wing Chaotic Attractor, its Electronic Circuit Designs and Microcontroller-Based Random Number Generator

An exhaustive analysis of a four-wing chaotic system is presented in this paper. It is proved that the evolution range of some variables can be modulated easily by one coefficient of a cross product term. An amplitude-adjustable chaotic circuit is designed, which shows a good agreement with the theoretical analysis. Also, in this paper a microcontroller-based random number generator (RNG) was designed with a nonlinear four-wing chaotic system. RNG studies of the current time have been usually carried out with complicated structures that are costly and difficult to use in real time implementations and that require so much energy consumption. On the other hand, in this paper, as opposed to the disadvantages mentioned here, a microcontroller-based RNG was designed with a four-wing chaotic system (also discussed in the paper) and this was introduced to literature. Microcontroller-based random numbers that passed randomness tests will be available for use in many fields in real life, particularly in encryption.

NON-ALCOHOLIC FATTY LIVER DISEASE AND CEREBRAL SMALL-VESSEL DISEASE IN COGNITIVELY NORMAL INDIVIDUALS

Many studies investigated the association between nonalcoholic fatty liver disease (NALFD) and cardiovascular diseases. However, the relationship between NAFLD and cerebral small vessel disease has not been thoroughly investigated yet. A total of 1283 participants from health promotion center cohort were included in analysis. The cerebral SVD burdens were assessed with white matter hyperintensities (WMH), lacunes, and microbleeds (MB). US diagnosis of fatty liver was made based on standard criteria, and NFS was used to assess severity of fibrosis and to classify participants with NAFLD in two groups: high-intermediate (NFS ≥ -1.455) and low probability (NFS < -1.455) of advanced fibrosis. Multivariable logistic regression was used to assess the association between NAFLD and severity of WMH, and presence of lacunes and MB after controlling for age, sex, smoking, alcohol, Body mass index, intracranial volume, systolic blood pressure, LDL cholesterol, use of lipid lowering medications, use of antidiabetic medications, and use of antihypertensive medications. The association of NAFLD with the severity of WMH was also explored in clinically relevant subgroups, and the interaction of NAFLD with clinical characteristics was tested using Wald tests for cross-product terms in regression models. NAFLD had a significant association with WMH severity but not with presence of lacunes nor MB. Multivariable logistic regression showed that compared to participants without NAFLD, participants with NAFLD had more frequent moderate to severe WMH. compared to participants without NAFLD, the OR for WMH in participants with low NFS and with intermediate to high NFS were 0.97 (0.57, 1.65) and 2.12 (1.44, 3.12), respectively. Especially, linear trend tests showed a significant trend for more severe WMH among non-NAFLD, NAFLD with low NFS and NAFLD with high NFS (P for trend <0.001). The positive association between NAFLD and WMH was prominent in the subgroups with middle age, female, low education, and non-obesity. Our findings suggested that NAFLD has a significant association with moderate to severe WMH in cognitively normal individuals and NAFLD severity also predicted more severe WMH. Odds ratios for prevalence of moderate or severe white matter hyperintensity by non-alcoholic fatty liver disease status in predefined subgroups.* Adjusted for age and sex, body mass index (underweight, normal, overweight, and obese), education (< middle school and ≥ middle school), smoking (never, former or current, and missing), and alcohol (none, and moderate), systolic blood pressure, fasting glucose, LDL cholesterol, use of diabetes medication use of antihyperlipidemia medication, and use of antihypertensive medication.

Abstract P110: Gene-Gene Interaction: SERPINE1 Interacts with ANRIL in Association with Coronary Artery Disease and Coronary Artery Calcium in a Non-Hispanic Population

Introduction: Long non-coding RNAs (lncRNA) are important regulators of gene expression. ANRIL ( CDKN2B-AS1 ) is an anti-sense lncRNA coded for at the Chr9p21 locus. GWAS have reported SNPs in the Chr9p21 to be associated with clinical coronary artery disease (CAD) and coronary artery calcium (CAC). A knockdown study of two isoforms of ANRIL reported the potential of ANRIL’s regulation of SERPINE1 expression. SERPINE1 encodes plasminogen activator inhibitor-1 (PAI-1) and has been reported to be associated with CAD and CAC. The 3’UTR of SERPINE1 has been reported to bind miRNA and lncRNA suggesting ANRIL may modify SERPINE1’s association with CAD and CAC. Hypothesis: We hypothesized that rs7242 or rs1050955, covering the LD of the 3’ UTR of SERPINE1, will interact with SNPs within ANRIL to affect their association with self-reported CAD and CAC in non-Hispanic Whites (NHW, n=6103) and African Americans (AA, n=2549) in the COPDGene Study. Methods: Interaction of SNPs from Illumina HumanOmniExpress Beadchip for ANRIL , and rs7242 and rs1050955 were evaluated in an additive genetic model with a cross-product interaction term. Logistic regression was used to evaluate the association of CAD (Yes/No) and CAC (Yes/No) with the SNP-SNP interactions, while controlling for genetic ancestry, ascertainment, age, sex, BMI, diabetes, hypertension, high cholesterol and pack-years of smoking. Analyses were stratified by race. Results: A significant interaction between rs7242 ( SERPINE1 ) and rs7049105 ( ANRIL ) and CAD (p=0.0004) and CAC (p=0.0023) were found in NHW (Figure 1). The risk estimates for the rs7242 G allele in SERPINE1 were higher in those with the rs7049105 AA genotype in ANRIL but were lower in those with the rs7049105 GG genotype in ANRIL . We did not find any significant interactions in AA. Conclusion: In this study, the previously identified GWAS signal for CAD and CAC in Chr9p21 ( ANRIL ) interacts with a SNP in SERPINE1 . This may be due to ANRIL lncRNA influences on gene expression of SERPINE1 and increases the risk of both CAC and CAD.

Linear Reformulation of Polynomial Discrete Programming for Fast Computation

Polynomial discrete programming problems are commonly faced but hard to solve. Treating the nonconvex cross-product terms is the key. State-of-the-art methods usually convert such a problem into a 0-1 mixed-integer linear programming problem and then adopt the branch-and-bound scheme to find an optimal solution. Much effort has been spent on reducing the required numbers of variables and linear constraints as well as on avoiding unbalanced branch-and-bound trees. This study presents a set of equations that linearize the discrete cross-product terms in an extremely effective manner. It is shown that embedding the proposed “equations for linearizing discrete products” into those state-of-the-art methods in the literature not only significantly reduces the required number of linear constraints from O(h3n3) to O(hn) for a cubic polynomial discrete program with n variables in h possible values but also tighten these methods with much more balanced branch-and-bound trees. Numerical experiments confirm a two-order (102-times) reduction in computational time for some randomly generated cubic polynomial discrete programming problems. There is a Video Overview associated with this article, available as supplemental material.

Amplitude modulation and its relation to streamwise convection velocity

The paper is devoted to show the relation between the amplitude modulation and the convection velocity in a turbulent boundary layer. The analysis has been performed based upon velocity profiles measured with the hot-wire technique in a turbulent boundary layer with pressure gradient corresponding to turbomachinery conditions. The VITA detection method of small-scale structures was used in order to show the impact of the pressure gradient on phase averaged time-traces changes in the aspect of amplitude modulation. The physical significance of the large scales influence on the small scales, formerly known as the amplitude modulation, has been elucidated. It was demonstrated that the measure of amplitude modulation can be used as an indicator of the convection velocity of coherent structures. The convection velocity profile can be described in the inner region using the amplitude modulation term when the proper scale is applied. The convection velocity prediction in favourable and adverse pressure gradients shows that t the strongest impact of the pressure gradient occurs in the buffer layer. The stronger increase occurs in the buffer layer. It was also found that the skewness factor of phase-averaged VITA waveforms, is very similar in shape with the cross-product term of the streamwise skewness factor, which quantifies amplitude modulation. It indicates that the well-known modification of quadrant events in pressure gradient flows is a result of the amplitude modulation phenomenon.

Complementary Log Regression for Sufficient-Cause Modeling of Epidemiologic Data.

The logistic regression model is the workhorse of epidemiological data analysis. The model helps to clarify the relationship between multiple exposures and a binary outcome. Logistic regression analysis is readily implemented using existing statistical software, and this has contributed to it becoming a routine procedure for epidemiologists. In this paper, the authors focus on a causal model which has recently received much attention from the epidemiologic community, namely, the sufficient-component cause model (causal-pie model). The authors show that the sufficient-component cause model is associated with a particular ‘link’ function: the complementary log link. In a complementary log regression, the exponentiated coefficient of a main-effect term corresponds to an adjusted ‘peril ratio’, and the coefficient of a cross-product term can be used directly to test for causal mechanistic interaction (sufficient-cause interaction). The authors provide detailed instructions on how to perform a complementary log regression using existing statistical software and use three datasets to illustrate the methodology. Complementary log regression is the model of choice for sufficient-cause analysis of binary outcomes. Its implementation is as easy as conventional logistic regression.

Relationship between Physical Disability and Depression by Gender: A Panel Regression Model.

BackgroundDepression in persons with physical disabilities may be more common than in the general population. The purpose of this study was to examine the relationship between physical disability and depression by gender among adults, using a large, nationally representative sample.MethodsThis study used data from the Korean Longitudinal Study of Aging, Wave one through four, and ran a series of random effect panel regression models to test the relationship between physical disability status and depression by gender. We tested the moderating effect of gender on the relationship between disability status and depression level by examining the significance of the cross-product term between disability status and gender.ResultsAfter controlling for self-rated health, marital status, employment status, education, and age, subjects who were female or diagnosed as having any disability presented higher levels of depression scores. Further, the difference in terms of their depression level measured by Center for Epidemiologic Studies Short Depression Scale (CES-D 10) scores between those who were diagnosed as having any disability and those who were not was greater for females than for their male counterparts.ConclusionThis study reaffirmed that disability is the risk factor of depression, using longitudinal data. In addition, female gender is the effect modifier rather than the risk factor. The effect of gender in the non-disability group, mostly composed of older persons, is limited. On the contrary, the female disability group showed more depressive symptoms than the male disability group. The gender difference in the disability group and the role of culture on these differences need further research.

Interaction of sex steroid hormones and obesity on insulin resistance and type 2 diabetes in men: the Third National Health and Nutrition Examination Survey

AimsWe examined interaction of sex steroid hormones and obesity with regard to insulin resistance (IR) and type 2 diabetes (T2D) by using nationally representative data from the US. MethodsData of 1461 men aged ≥20years who participated in the Third National Health and Nutrition Examination Survey were analyzed. Multiplicative interaction was calculated by cross-product interaction terms in multivariable logistic regression models. Additive interaction was assessed by the relative excess risk due to interaction (RERI). ResultsAfter adjusting for demographic and lifestyle covariates, the odds of IR were greatest among obese men with low free testosterone and high androstanediol glucuronide. Multiplicative interactions for total testosterone, free testosterone, and free estradiol index (FEI) were statistically significant with central obesity but not with overweight and obesity regarding to T2D (P<0.05). Significant additive interactions with obesity or central obesity were detected for total testosterone (RERI=2.75, 95% CI=0.92,4.59), SHBG (RERI=5.71, 95% CI=0.77,10.64), and FEI (RERI=−9.96, 95% CI=−19.18,-0.74) with regard to IR, beta-cell dysfunction, and T2D. ConclusionsOur findings add to the evidence suggesting that low testosterone and high estradiol may be associated greater risks of IR and T2D by interacting with overall and central obesity in adult men.

Breaking time reversal symmetry with coriolis mean flow systems

One way to break time reversal symmetry is to have a moving wave propagation medium. If the acoustic wave vector and the moving fluid velocity are collinear, such as the one created by Fluery et al. (Science, 343, 2014) we can use the wave vector shift caused by the fluid flow to break reciprocity. An alternative approach we have taken is to use a fluid velocity field which enters the differential equation of the system as a cross product term with the wave vector. A circular field in which the fluid velocity increases radially has a Coriolis acceleration term. In this system, the acoustic wave enters from the central wall and exits from the perimeter wall. Equations of conservation of mass and momentum, after linearization can be simplified to: (∂/∂t+V.▽)P+ρC2▽.(u+V) = 0, ρ∂/∂t(u+V)+ρ(V.▽)u+ρ(u.▽)V+ρ(V.▽)V+▽P = 0, in which u is the particle acoustic velocity and V = rωeθis the fluid velocity due to the rotational velocity field. It can be shown that the third term in the second equation can be simplified as:ρωek×u. This term Coriolis acceleration induces nonreciprocity. In this work, we solved the differential equation numerically and investigated the effect of fluid velocity on the nonreciprocity factor.

Amplitude-Phase Modulation, Topological Horseshoe and Scaling Attractor of a Dynamical System**Supported by Hunan Provincial Natural Science Foundation of China under Grant No. 2016JJ4036, University Natural Science Foundation of Jiangsu Province under Grant No. 14KJB120007 and the National Natural Science Foundation of China under Grant Nos. 11504176 and 11602084

A three-dimensional autonomous chaotic system is discussed in this paper. Some basic dynamical properties of the system, including phase portrait, Poincaré map, power spectrum, Kaplan–Yorke dimension, Lyapunov exponent spectra, signal amplitude and topological horseshoe are studied theoretically and numerically. The main finding by analysis is that the signal amplitude can be modulated via controlling the coefficients of the linear term, cross-product term and squared term simultaneously or respectively, and the phase of x3 can be modulated by the product of the coefficients of the linear term and cross-product term. Furthermore, scaling chaotic attractors of this system are achieved by modified projective synchronization with an optimization-based linear coupling method, which is safer for secure communications than the existed synchronization scheme since the scaling factors can be regarded as the security encoding key.

Improved stability criteria for generalized neural networks with time-varying delay by auxiliary function-based integral inequality

This paper is mainly concerned with improved stability criteria for generalized neural networks (GNNs) with time-varying delay by delay-partitioning approach. A newly augmented Lyapunov-Krasovskii functional (LKF) with triple integral terms is constructed by decomposing integral interval, in which the relationships between the augmented state vectors are fully taken into account. The tighter bounding inequalities such as a Wirtinger-based integral inequality, Peng-Park’s integral inequality, and an auxiliary function-based integral inequality are employed to effectively handle the cross-product terms occurred in derivative of the LKF. As a result, less conservative delay-dependent stability criterion can be achieved in terms of $e_{s}$ and LMIs. Finally, two numerical examples are included to show that the proposed results are less conservative than existing ones.

The association of dopamine pathway gene score, nicotine dependence and smoking cessation in a rural male population of Shandong, China.

Smoking and smoking cessation are both psychological and physiological traits. We aimed to investigate the interaction between dopamine pathway gene scores and nicotine dependence on smoking cessation in a rural Chinese male population. Participants were recruited from 17 villages in Shandong, China. DNA was extracted from blood sample of 819 participants. 25 single nucleotide polymorphisms (SNPs) in 8 dopamine (DA) pathway genes were genotyped. Weighted gene score of each gene is used to analyze the whole gene effect. Logistic regression was used to calculate odds ratios (OR) and multivariate-adjusted OR of each gene score for smoking cessation. Multiplicative model interaction was assessed through a cross-product interaction term of gene score by nicotine dependence in a multivariate logistic regression model. After adjusting for age, occupation, education, marital status, self-rating anxiety score, and disease status, we observed significant negative associations of catechol-O-methyltransferase (COMT), dopamine receptor D2 (DRD2) gene score and smoking cessation, as well as significant positive associations between ankyrin repeat and kinase domain containing 1 (ANKK1), dopamine transporter (SLC6A3), dopamine receptor D4 (DRD4) gene score and smoking cessation. A significant multiplicative model interaction between nicotine dependence and the SLC6A3 gene score on smoking cessation was also observed (p = .03). There is a significant multiplicative model interaction of SLC6A3 gene score and nicotine dependence on smoking cessation. This finding could help to identify smokers who may be at high risk of relapse, and thus to develop more professional and personalized smoking cessation treatment. (Am J Addict 2016;25:493-498).