Influence maximization, defined by Kempe, Kleinberg, and Tardos (2003), is the problem of finding a small set of seed nodes in a social network that maximizes the spread of influence under certain influence cascade models. In this paper, we propose an extension to the independent cascade model that incorporates the emergence and propagation of negative opinions. The new model has an explicit parameter called quality factor to model the natural behavior of people turning negative to a product due to product defects. Our model incorporates negativity bias (negative opinions usually dominate over positive opinions) commonly acknowledged in the social psychology literature. The model maintains some nice properties such as submodularity, which allows a greedy approximation algorithm for maximizing positive influence within a ratio of 1 − 1/e. We define a quality sensitivity ratio (qs-ratio) of influence graphs and show a tight bound of Θ( n/k) on the qs-ratio, where n is the number of nodes in the network and k is the number of seeds selected, which indicates that seed selection is sensitive to the quality factor for general graphs. We design an efficient algorithm to com- * Author affiliations and emails: W. pute influence in tree structures, which is nontrivial due to the negativity bias in the model. We use this algorithm as the core to build a heuristic algorithm for influence maximization for general graphs. Through simulations, we show that our heuristic algorithm has matching influence with a standard greedy approximation algorithm while being orders of magnitude faster.

Antibodies elicited by protein therapeutics can cause serious side effects in humans. We studied immunogenicity of a recombinant fusion protein (FPX) consisting of two identical, biologically active, peptides attached to human Fc fragment. EpiMatrix, an in silico epitope-mapping tool, predicted promiscuous T-cell epitope(s) within the 14-amino-acid carboxy-terminal region of the peptide portion of FPX. On administration of FPX in 76 healthy human subjects, 37% developed antibodies after a single injection. A memory T-cell response against the above carboxy-terminus of the peptide was observed in antibody-positive but not in antibody-negative subjects. Promiscuity of the predicted T-cell epitope(s) was confirmed by representation of all common HLA alleles in antibody-positive subjects. As predicted by EpiMatrix, HLA haplotype DRB1*0701/1501 was associated with the highest T-cell and antibody response. In conclusion, in silico prediction can be successfully used to identify Class II restricted T-cell epitopes within therapeutic proteins and predict immunogenicity thereof in humans.

No abstract

No abstract

No abstract

Objective-We examined whether B-mode ultrasound-detected carotid artery intima-media thickness (IMT) was elevated before the onset of clinical diabetes. Methods and Results-The study population for these analyses included 1127 nondiabetic participants, 66 prediabetic participants, and 303 diabetic participants with a mean age of 49.8 years who participated in the Mexico City Diabetes Study, a prospective cohort study. A lthough previous studies have established that coronary heart disease (CHD) risk factors are elevated before the clinical onset of diabetes 1-5 and a recent study has reported increased cardiovascular disease risk before the clinical onset of diabetes, 6 direct evidence of increased atherosclerosis before a clinical diagnosis of diabetes has not been documented. Documenting not only increased cardiovascular risk factors but increased atherosclerosis as well provides direct evidence that the atherosclerotic process is accelerated before the onset of clinical diabetes. Moreover, although type 2 diabetes is commonly considered a risk factor for CHD, treatment and control of hyperglycemia in type 2 diabetes have, at best, only a modest effect on reducing the risk of CHD associated with diabetes. 7 If the atherosclerotic process is accelerated before the onset of clinical diabetes, then preventing diabetes itself, either by altering lifestyle or pharmacologically, might be the optimum way to reduce the CHD risk associated with diabetes. See page 1715Recently, the Diabetes Prevention Program was one of several clinical trials to indicate that either through diet and exercise or with the aid of a pharmacologic agent it is possible to lower the incidence of diabetes among individuals at high risk for the disease. 8 -10 In short, there are lifestyle and pharmacologic interventions available that prevent diabetes. Hence, evidence suggesting that prediabetic individuals have not only elevated CHD risk factors but also elevated levels of subclinical atherosclerosis would indicate the importance of a lifestyle or pharmacologic intervention before the onset of clinical diabetes.Therefore, we examined (1) whether B-mode ultrasounddetected carotid artery intima-media thickness (IMT) was elevated in prediabetic individuals when compared with individuals who remained free of diabetes and (2) whether increased carotid artery IMT predicted incident diabetes. Methods The Mexico City Diabetes StudyThe Mexico City Diabetes Study is a population-based cohort of 2282 men and women first examined between 1990 and 1992 who were invited to return for 2 follow-up exams, the first conducted between 1993 and 1995 and the second from 1997 to 1999. 11 Participants were randomly selected from 6 low-income "colonias"

B. thuringiensis cells evolve from vegetative cells to sporulated cells during batch growth. As a result, the classical model based on an exponential binary fission and the Monod equation has intrinsic limitations to describe the biomass concentration. A new kinetic model accounting for the spore formation process is presented in this study. This model considers that only the cells without a spore are able to contribute to the cell growth. This model also incorporates the spore formation process using a spore formation step and a specific spore formation rate constant. Classical and new model predictions are compared with batch experimental data. Results demonstrate that the classical model is unable to predict the experimental data and this is particularly true from the middle of the transition stage on. In contrast, the new sporulation kinetic model is able to predict the experimental data more accurately for the complete time span of the batch culture.Les cellules B. thuringiensis passent de I'etat de cellules vegetatives a I'etat de cellules sporulees pendant une croissance discontinue. Par consequent, le modele classique reposant sur une fission binaire exponentielle et I'equation de Monod presentent des limitations intrinseques pour decrire la concentration de biomasse. On presente dans cette etude un nouveau modele tenant compte du processus de formation des spores. Ce modele considere que seules les cellules sans spore sont capables de contribuer a la croissance des cellules. Ce modele incorpore egalement le processus de formation des spores en recourant a une etape de formation des spores et a une constante de vitesse de formation des spores specifique. Les predictions du modele classique et de ce nouveau modele sont comparees a des donnees experimentales discontinues. Les resultats demontrent que le modele classique est incapable de predire les donnees experimentales, ce qui est particulierement vrai a partir du milieu de la phase de transition. A I'inverse, le nouveau modele de sporulation cinetique est capable de predire les donnees experimentales avec davantage de precision pour la duree complete de la culture discontinue.

No abstract