Gametocytes infectiousness to mosquitoes: variable selection using random forests, and zero inflated models

نویسندگان

  • Robin Genuer
  • Isabelle Morlais
  • Wilson Toussile
چکیده

Malaria control strategies aiming at reducing disease transmission intensity may impact both oocyst intensity and infection prevalence in the mosquito vector. Thus far, mathematical models failed to identify a clear relationship between Plasmodium gametocytes and their infectiousness to mosquitoes. Natural isolates of gametocytes are genetically diverse and biologically complex. Infectiousness to mosquitoes relies on multiple parameters such as density, sexratio, maturity, parasite genotypes and host immune factors. In this article, we investigated how density and genetic diversity of gametocytes impact on the success of transmission through the mosquito vector. We analyzed data for which the number of variables plus attendant interactions is at least of order of the sample size, precluding usage of classical models such as general linear models. We then applied a variable selection procedure based on the random forests score of variable importance. The selected variables were assessed in the zero inflated negative binomial model which accommodates both over-dispersion and the sources of non infected mosquitoes. We found that the most important variables related to infection prevalence and parasite intensity are gametocyte density and multiplicity of infection. Key-words: Plasmodium, mosquitoes, variable selection, random forests, zero inflated models. ∗ Université Paris-Sud, Laboratoire de Mathématique, UMR 8628, Orsay cedex F-91405 † Inria Saclay Ile-de-France ‡ UR016, Institut de Recherche Pour le Développement, 911 Avenue Agropolis, PO Box 64501, F-34394 Montpellier Cedex 5 in ria -0 05 50 98 0, v er si on 3 21 F eb 2 01 1 Capacité d’infection des gamétocytes aux moustiques : sélection de variables basée sur les forêts aléatoires, et modèles modifiés en zéro Résumé : De nouvelles stratégies de réduction de la transmission du paludisme nécessite la compréhension des facteurs pouvant influencer l’intensité d’oocystes et la prévalence d’infection chez le moustique vecteur. Jusqu’à maintenant, les modèles mathématiques ne sont pas parvenus à identifier une relation claire entre les gamétocytes de Plasmodium et leur capacité à infecter les moustiques. La capacité vectorielle du moustique peut dépendre de multiple facteurs tels que la densité, le sexe-ratio et la maturité du parasite, ainsi que des facteurs immunitaires du moustique. Dans ce papier, nous évaluons l’influence de la densité et de la diversité génétique du parasite sur le succès de sa transmission à travers le moustique vecteur. Nous disposons de données décrites par diverses variables dont le nombre est de l’ordre de la taille de l’échantillon, ce qui constitue un obstacle à l’usage de modèles classiques de régression tel que le modèle linéaire généralisé. Nous considérons alors l’importance des variables des forêts aléatoires pour sélectionner les variables les plus influentes. Les variables sélectionnées sont ensuite évaluées par le modèle binomial négatif modifié en zéro, qui permet de tenir compte à la fois de la sur-dispersion et des sources possibles des moustiques non-infectés. Nous trouvons que les variables les plus importantes reliées à la prévalence d’infection et l’intensité parasitaire sont la densité de gamétocytes et la multiplicité de l’infection. Mots-clés : Plasmodium, moustiques, sélection de variables, forêts aléatoires, modèles modifiés en zéro. in ria -0 05 50 98 0, v er si on 3 21 F eb 2 01 1 Gametocytes infectiousness to mosquitoes using random forests 3

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Zero inflated Poisson and negative binomial regression models: application in education

Background: The number of failed courses and semesters in students are indicatorsof their performance. These amounts have zero inflated (ZI) distributions. Using ZI Poisson and negative binomial distributions we can model these count data to find the associated factors and estimate the parameters. This study aims at to investigate the important factors related to the educational performance of ...

متن کامل

Assessment of length of stay in a general surgical unit using a zero-inflated generalized Poisson regression

Background: The effective use of limited health care resources is of prime importance. Assessing the length of stay (LOS) is especially important in organizing hospital services and health system. This study was conducted to identify predictors of LOS among patients who were admitted to a general surgical unit.    Methods: In this cross-sectional study, the sample included all patien...

متن کامل

A protocol for membrane feeding assays to determine the infectiousness of P. falciparum naturally infected individuals to Anopheles gambiae

This protocol describes procedures to conduct membrane feeding assays to assess the infectiousness of individuals living in malaria endemic areas. Humans may be infectious to mosquitoes if they have mature male and female gametocytes in their peripheral blood. Because of the limited sensitivity of microscopy to detect gametocytes, epidemiological studies that aim to enroll all potentially infec...

متن کامل

Hurdle, Inflated Poisson and Inflated Negative Binomial Regression Models ‎ for Analysis of Count Data with Extra Zeros

In this paper‎, ‎we ‎propose ‎Hurdle regression models for analysing count responses with extra zeros‎. A method of estimating maximum likelihood is used to estimate model parameters. The application of the proposed model is presented in insurance dataset‎. In this example‎, there are many numbers of claims equal to zero is considered that clarify the application of the model with a zero-inflat...

متن کامل

Infectiousness of the human population to Anopheles arabiensis by direct skin feeding in an area hypoendemic for malaria in Senegal.

Direct skin feeding experiments are sensitive assays to determine human infectiousness to mosquitoes but are rarely used in malaria epidemiological surveys. We determined the infectiousness of inhabitants of a malaria hypoendemic area in Senegal. Gametocyte prevalence by microscopy was 13.5% (26 of 192). Of all individuals who were gametocyte positive, 44.4% (11 of 25) infected ≥ 1 Anopheles ar...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2011