HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Varona, L. Articles by Noguera, J. L. Search for Related Content PubMed PubMed Citation Articles by Varona, L. Articles by Noguera, J. L.

Bayes factor analysis for the genetic background of physiological and vitality variables of F₂ Iberian x Meishan newborn piglets¹

L. Varona^*,2,3, J. Casellas^,3, J. Piedrafita, A. Sánchez, P. Garcia-Casado, M. Arqué^* and J. L. Noguera^*

^* Area de Producció Animal, Centre UdL-IRTA, 25198 Lleida, Spain; and Departament de Ciència Animal i dels Aliments, Universitat Autònoma de Barcelona, 08193 Bellaterra (Barcelona), Spain; and and Dto. Reproducción Animal, Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, 28080, Madrid, Spain

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

 
The Bayes factor (BF) procedure was applied to examine the additive genetic component of several physiological and vitality variables for newborn pigs. Nine variables were studied: heart rate, arterial oxygen saturation, rectal temperature (all at birth and 60 min later), birth weight, interval between birth and first teats contact, and interval between birth and first colostrum intake. The available numbers of data ranged from 288 (heart rate at 60 min) to 839 records (birth weight) from F₂ Iberian x Meishan newborn pigs. We compared a model with zero heritability (nonheritable) with the one where the additive genetic background was included. The BF was used to discriminate between both candidate models. Very strong evidence of genetic background was detected for heart rate 60 min after birth (BF = 48.90), and strong evidence was detected for rectal temperature at birth (BF = 13.82). Posterior modes (means) of heritabilities were 0.29 (0.32) and 0.40 (0.39), respectively. In addition, substantial evidence of absence of genetic background was detected for arterial oxygen saturation at birth.

Key Words: Bayes Factor • Neonate Metabolism • Survival • Vitality

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

 
Preweaning mortality represents a significant economic loss in pig production, although selection of this trait has not been considered because of its low heritability (Knol et al., 2002). Therefore, indirect genetic improvement through related traits such as birth weight or the within-litter homogeneity in birth weight has been proposed (Knol et al., 2002).

Early physiological or behavioral variables may also represent an alternative approach by predicting the survival ability of the piglets. After birth, newborn pigs must adapt to the new environment quickly because a delay could create a great disadvantage in terms of access to resources (Fraser and Thompson, 1991). In addition, Herpin et al. (1996, 1998) reported that the degree of asphyxiation suffered during delivery had a significant influence on piglet adaptation and posterior viability. This period of hypoxia also initiates drastic changes in various physiological variables that, which in turn can influence piglet survivability (Herpin et al., 1996, 1998; Tuchscherer et al., 2000). Several authors have related preweaning piglet mortality with early metabolic variables such as body temperature or arterial oxygen saturation (Herpin and Le Dividich, 1998, Herpin et al., 2001; Casellas et al., 2004a), or initial suckling behavior (Herpin et al., 1996; Tuchscherer et al., 2000). Moreover, heart rate has been also included in the vitality score (Randall, 1971).

However, knowledge about the additive genetic variation of these traits is practically nonexistent. One important question is whether there is a genetic background in the observed phenotypic variation. Recently, a Bayes factor (BF) procedure for testing the genetic background of quantitative traits was developed (García-Cortés et al., 2001; Varona et al., 2001). In this context, this article is an attempt to examine the additive genetic background of several physiological and vitality variables closely related to the first minutes of life after birth.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

 
Animals
One hundred eight litters from Iberian x Meishan F₁ gilts mated with eight Iberian x Meishan F₁ boars were recorded in Nova Genètica farm of Solsona (Lleida, Spain) between November 2001 and May 2002. These F₁ individuals came from three unrelated Iberian boars and 18 Meishan sows, and they provided 1,039 piglets born alive, 34 stillbirths, and 27 mummified fetuses. Matings were designed to avoid possible inbreeding. These data were obtained in the scope of a F₂ design for detection of QTL for reproductive traits (Grant AGL2000-1229-C03). Primiparous sows were housed in climate-controlled rooms (24°C) during farrowing and lactation, and penned in standard farrowing crates with heating plates for piglets (38°C). Feeding of sows was restricted during the gestation period (9.2 MJ of NE, 13.5% CP, and 0.48% lysine), and ad libitum during lactation (9.8 MJ of NE, 17.5% CP, and 0.82% lysine). During parturition, veterinary interventions were kept to a minimum; 20 I.U. of oxytocin was administered by i.m. injection when the interval between births exceeded 30 min. Furthermore, sows that did not deliver before d 113 of gestation received 175 µg of cloprostenol to elicit farrowing (intravulvar injection).

Animal procedures followed ethical norms of IRTA.

Physiological and Vitality Trait Recording
Birth weight, time, and presentation (anterior or posterior) were recorded for each piglet, and the vitality score was determined following in part the procedures described by Randall (1971). Three variables were scored by subjective evaluation (see Casellas et al., 2004b): time to the onset of respiration, muscle tone, and attempts to stand. The sum of these three values defined vitality score. Piglets were monitored for heart rate (HRL), arterial oxygen saturation (OS1), and rectal temperature at birth (RT1) and 60 min later (HR2, OS2, RT2), using a Vet/Ox 4404 monitor pulsoxymeter (Heska Corp., Fort Collins, CO). Given that all the piglets had black skin, heart rate and arterial oxygen saturation measurements were taken by tongue, using an Exotic Reflectance Sensor (Heska Corp.). The time interval between birth and the measurement of OS was also recorded, which was used to adjust arterial oxygen saturation (OS) to 90 s after birth (OS1), applying the GLM procedure of SAS (SAS Inst., Inc., Cary, NC) because Herpin et al. (1998) observed a large increase in OS during the first seconds of life. All piglets were ear tagged for identification purposes, weighed, and placed near the posterior side of their mother. This allowed for recording the time that elapsed between birth and first teat contact, as well as the first suckle (TS).

Bayes Factor Analysis
For each variable the significant effects previously analyzed in Casellas et al. (2004b) were considered (Table 1), with the exception of birth weight (BWT), which was analyzed following in part the fixed model proposed by Roehe (1999), with sex and litter size as systematic effects, and a random source of variation characterized by the litter. We estimated the BF between the models with and without genetic effects following García-Cortés et al. (2001) and Varona et al. (2001). The first model (Model 1) includes the additive genetic effects:

View this table: [in this window] [in a new window]   Table 1. Number of individuals studied, means (), and standard errors of the mean (SEM) for the analyzed variables, as well as the fixed and significant random effects included in mixed models

where y contains phenotypic records; X, Z₁, and Z₂ are the incidence matrices relating observations to systematic (ß), additive genetic (u), and random litter (p) effects; and e is the vector of residuals. The vectors u, p, and e are assumed to be normally distributed.

where ²_u is the additive genetic variance, ²_p is the variance of the random litter effects, ²_e is the residual variance, and A is the numerator relationship matrix. Model 1 can be reparameterized as:

where

Consequentsly,

where ² = ²_u/²_t is the intraclass correlation, and ²_t = (²_u + ²_e) is the variance of e*. The joint distribution of all variables in Model 1 is:

where

and 0 otherwise for each level i of ß)

where k₁, k₂, and k₃ are three small enough values to ensure a flat distribution over the parametric space.

The alternative model (Model 2) is:

where

Then, the joint distribution of records and parameters is:

where we can assume that prior distributions p₂(ß), p₂ (p |²_p), p₂(²_p), and p₂(²_t) are identical to the prior distributions of the previous model. And the likelihood of the model is

According to Varona et al. (2001), only the analysis with the complex model (Model 1) is required. Following García-Cortés et al. (2001) and Varona et al. (2001):

because p₁(² = 0) = 1 or simultaneously:

A BF₁₂ (BF₂₁) sss>1 (<1) indicates that the model with additive variance components (heritable) is more suitable. On the contrary, a BF₁₂ (BF₂₁) <1 (>1) indicates that the non heritable model is more probable. From now on, we will refer to BF₁₂ as BF.

According to Jeffreys (1984), the BF can be classified according to the levels of evidence:

BF <1 Null hypothesis supported 1 < BF < 3.16 Not worth more than a bare mention 3.16 < BF < 10 Substantial evidence 10 < BF < 31.62 Strong evidence 31.62 < BF < 100 Very strong evidence BF >100 Decisive

Sampling from the conditional distribution of ² was performed using a Gibbs sampler (Gelfand and Smith, 1990), with a Metropolis-Hastings step (Hastings, 1970). To calculate the posterior distribution of the heritability (h²), the following formula was used every iteration:

A total of 12,500 iterations were performed, and 10,000 iterations were used after discarding the first 2,500. All correlated samples were used to calculate the posterior distributions using the ergodic property of the chain (Gilks et al., 1996). Convergence was checked using the algorithm of Raftery and Lewis (1996).

	Results

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

 
A variable number of records was analyzed for each variable, ranging from 288 (HR2) and 869 (BWT) newborn pigs. The average BWT was 1.14 kg, with a SD of 0.24 kg. Sex and the number of piglets and random litter effects were included in the assumed model for BWT. In addition, Table 1 shows average values, standard errors, and the assumed model for each vitality and physiological variable. The assumed model for each variable was determined by the results obtained by Casellas et al. (2004b) with the same data set. A detailed discussion of these results was presented in that study.

Five variables reached average BF larger than 1: heart rate 60 min after birth (48.90), rectal temperature at birth (13.82), birth weight (2.27), heart rate at birth (1.22), and the interval between birth and first teat contact (1.03). Following the classification of the BF by Jeffreys (1984), there was very strong evidence for the presence of genetic background of HR2, and strong evidence in RT1. The posterior modes (means in parentheses) of their heritability were 0.13 (0.22), 0.21 (0.38), 0.29 (0.32), 0.40 (0.39), and 0.17 (0.38) for BWT, heart rate at birth and 60 min later, RT1, and time between birth and first teat contact, respectively (Table 2). The rest of the variables attained BF <1, varying between 0.56 for elapsed time between birth and first suckle and 0.29 for OS1, all with posterior modes for heritability smaller than 0.13 (Table 2). The distributions of the heritability estimates were clearly asymmetrical, revealing great differences between the posterior mode and the posterior mean for the heritability, with the exception of the HR2 and RT1 variables.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

The application of the BF for testing the genetic background described by García-Cortés et al. (2001) requires the reparameterization of the model in terms of phenotypic variance and intraclass correlation because the BF can only be computed between nested models that differ in a bounded variable using this approach. In this case, the bounded variable is the intraclass correlation, and the comparison between models is calculated between a model with zero intraclass correlation (nonheri-table) against a model with intraclass correlation between 0 and 1 (heritable).

The main disadvantage of the application of the BF is its strong dependence on the assumed prior distributions. However, in this case, the prior distributions for all parameters, with the only exception of the intraclass correlation, are cancelled because they are assumed to be the same in both candidate models (Varona et al., 2001). In addition, the final formula only includes the marginal prior and posterior distribution for the in-traclass correlation (²), and, as a consequence, the BF results are robust for modifications of this prior distribution. When the prior distribution of ² is modified, the posterior distribution changes in the same direction and, as a consequence, the BF remains stable.

Physiological variables and initial feeding behavior may play an important role during adaptation to the external environment and subsequent survivability in newborn pigs (Fraser and Thompson, 1991; Herpin et al., 1998; Casellas et al., 2004a). Variables like heart rate, rectal temperature, or OS characterize the metabolic situation of piglets after birth, the degree of asphyxiation suffered during delivery, and the ability to recover physiological levels (Herpin et al., 1996, 1998; Casellas et al., 2004b). Notwithstanding, our knowledge of these variables is limited, especially concerning their genetic component. To our knowledge, this is the first study that analyzed the genetic background of heart rate, arterial oxygen saturation, or rectal temperature in neonates. Thus, our results open a new research field related to neonate physiology in general and the genetic aspects of newborn pig metabolism in particular.

Rectal temperature at birth or during the first day of life is historically the best-studied variable among those treated in this investigation (see Herpin and Le Dividich, 1998), perhaps with the exception of birth weight, but our knowledge about the possible heritability of body temperature is restricted to adult individuals. In this context, the evidence of heritable variation for body temperature has been the object of an important discussion in recent decades. Silcock and Parsons (1973) observed an additive genetic component for body temperature in mice, but subsequent studies were unable to find significant heritabilities (Lacy and Lynch, 1978; Lynch et al., 1988). Recently, a response of rectal temperature to selection was demonstrated in laboratory rats (Gordon and Rezvani, 2001) but, on the other hand, the heritability estimates for rectal temperature in leaf-eared mice (Phyllotis darwini) did not reach the statistical threshold, although other variables closely related to animal thermoregulation, such as thermal conductance, showed significant heritabilities (Nespolo et al., 2003). Unfortunately, our results did not allow us to clarify this controversy. Although rectal temperature at birth showed a large BF (13.82), with a moderate to high posterior mode for the heritability (0.40), rectal temperature 60 min after birth reached a lower posterior mode (0.11) and BF (0.42). This implies a great change in heritability over a short temporal space, although Casellas et al. (2004b) found a significant phenotypic correlation between both measures in the same data set. This results are perhaps comparable with the ones obtained in leaf-eared mice, where diurnal temperature was clearly nonheritable, whereas the genetic component of nocturnal temperature was almost significant (P < 0.10; Nespolo et al., 2003).

Moderate heritabilities were estimated for heart rate variables, with a BF >1, especially for heart rate 60 min after birth (48.90). In this context, an analysis carried out in adult humans also described a moderate heritability for heart rate (Singh et al., 1999), and Barrey (1992) reported a high heritability for the maximum heart rate in adult human athletes (0.72). These results indicate the presence of a genetic component for these variables, compatible with the description of a QTL associated with heart rate in an F₂ experiment involving CBA/CaJ and BALB/cJ mice strains (Sugiyama et al., 2002). However, both heart rate variables we measured have to be considered as different traits because Casellas et al. (2004b) did not find a significant genetic correlation.

Bayes factors for arterial oxygen saturations did not reach values greater than 1, so there is no evidence for the existence of an additive genetic component for these variables in neonates. It is also remarkable that the BF for OS1 was 0.29, or 3.45 in the opposite sense; thus, there is substantial evidence in favor of the model without genetic determinism. Notwithstanding, a major gene for OS has been described in Tibetan highlanders, obtaining a heritability of 0.41 ± 0.14 (Beall et al., 1994). Unfortunately, the effect of this major gene has never been evaluated in neonates or in swine. Similar values were obtained for the interval between birth and the first teats contact, as well as the interval between birth and the first colostrum intake. Only the interval between birth and the first teat contact reached a BF >1, but very close to 1. This result implies that the time spent searching for teats may be influenced by an additive component, but further studies are necessary to confirm this hypothesis.

Birth weight reached a moderate heritability (0.13) with a BF of 2.27. It did not provide us with substantial evidence regarding the genetic background of this trait; the posterior mode for heritability was similar to the value reported by Kerr and Cameron (1995), and slightly greater than the one described by Roehe (1999). However, the results of this study are limited by the structure of the data set, given that the experiment was designed for QTL detection in reproductive traits. The F₂ design is not an appropriate design for detecting genetic background because in the F₂ generation, genetic components related to dominance and linkage disequilibrium between loci usually appear. The additive genetic variance that we were able to detect in this study was composed of half the genetic variance of the parental populations plus the segregation variance. This latter variance is the one that we shall be able to exploit in QTL analysis to detect specific loci related to the differences in these behavioral and physiological variables. Unfortunately, data from F₀ and F₁ generations were not available to allow for the discrimination between additive variances from the parental populations and the segregation variance (Birchmeier et al., 2002). In addition, it is expected that half the heterosis is still present in this F₂ generation and, in our opinion, its effect will be absorbed into the systematic effects.

	Implications

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

 
Strong evidence of genetic determinism of rectal temperature at birth and heart rate 60 min after birth was found in F₂ Iberian x Meishan newborn pigs. The genetic determinism of these traits suggests the possibility of using indirect selection to improve the survivability of piglets during the first days of life. The measurement of these traits is difficult, and as a consequence, their use in current breeding schemes in not feasible. Nevertheless, the genetic determinism of some of these traits involves the possibility of quantitative trait loci detection, which could be used in marker-assisted selection or introgression.

	Footnotes

 
¹ Financial support was provided by MCYT, Spain (Grant AGL2000-1229-C03). The authors are indebted to the staff of Nova Genètica for cooperating in the experimental protocol, in particular to F. Márquez, I. Riart, R. Malé, F. Rovira, M. González, and E. Ramells, and to M. Fina and J. Tarrés for their active and friendly collaboration during data collection. The authors are also indebted to J. P. Bidanel and J. C. Caritez (INRA) for their valuable help and L. Silió, M. C. Rodríguez, and C. Ovilo (INIA) for their cooperation. The authors gratefully acknowledge the contributions of the INRA (France) and the SIA El Dehesón del Encinar (Spain) for providing the purebred Meishan sows and Iberian boars, respectively. The English revision by C. Simmons is also acknowledged.

³ Both authors contributed equally to this manuscript.

² Correspondence—phone: 34 973 702637; e-mail: Luis.Varona{at}irta.es.

Received for publication October 8, 2004. Accepted for publication October 27, 2004.

	Literature Cited

Top Abstract Introduction Materials and Methods Results Discussion Implications Literature Cited

 


Barrey, E. 1992 Evaluation de l’aptitude sportive chez le cheval: Application à la définition de critères précoces de sélection. INRA Prod. Anim. Hors Série 167–173.

Beall, C. M., J. Blangero, S. Williams-Blangero, and M. C. Goldstein. 1994. Major gene for percent of oxygen saturation of arterial hemoglobin in Tibetan highlanders. Am. J. Phys. Anthrop. 95:271–276.[Medline]

Birchmeier, A. N., R. J. C. Cantet, R. L. Fernando, C. A. Morris, F. Holgado, A. Jara, and M. G. Santos-Cristal. 2002. Estimation of segregation variance for birth weight in beef cattle. Livest. Prod. Sci. 76:27–35.

Casellas, J., J. L. Noguera, L. Varona, A. Sánchez, M. Arqué, and J. Piedrafita. 2004a. Viability of Iberian x Meishan F₂ newborn pigs. II. Survival analysis up to weaning. J. Anim. Sci. 82:1925–1930.[Abstract/Free Full Text]

Casellas, J., W. M. Rauw, J. Piedrafita, A. Sánchez, M. Arqué, and J. L. Noguera. 2004b. Viability of Iberian x Meishan F₂ newborn pigs. I. Analysis of physiological and vitality variables. J. Anim. Sci. 82:1919–1924.[Abstract/Free Full Text]

Fraser, D., and B. K. Thompson. 1991. Armed sibling rivalry among suckling piglets. Behav. Ecol. Sociobiol. 29:9–15.

García-Cortés, L. A., C. Cabrillo, C. Moreno, and L. Varona. 2001. Hypothesis testing for the genetic background of quantitative traits. Genet. Sel. Evol. 33:3–16.[Medline]

Gelfand, A., and A. F. M. Smith. 1990. Sampling based approaches to calculating marginal densities. J. Am. Stat. Assoc. 85:398–409.

Gilks, W. R., S. Richardson, and D. J. Speigelhalter. 1996. Markov Chain Monte Carlo in Practice. Chapman and Hall, London, U.K.

Gordon, C. J., and A. H. Rezvani. 2001. Genetic selection of rats with high and low body temperatures. J. Therm. Biol. 26:223–229.[Medline]

Hastings, W. K. 1970. Monte Carlo sampling methods using Markov Chains and their applications. Biometrika 57:97–109.[Abstract/Free Full Text]

Herpin, P., J. C. Hulin, J. Le Dividich, and M. Fillaut. 2001. Effect of oxygen inhalation at birth on the reduction of early postnatal mortality in pigs. J. Anim. Sci. 79:5–10.[Abstract/Free Full Text]

Herpin, P., and J. Le Dividich. 1998. Termorregulación y entorno. Pages 57–97 in El Lechón Recién Nacido. M. A. Varley, ed. Acribia, Zaragoza, Spain.

Herpin, P., J. Le Dividich, J. C. Hulin, M. Fillaut, F. De Marco, and R. Bertin. 1996. Effects of the level of asphyxia during delivery on viability at birth and early postnatal vitality of newborn piglets. J. Anim. Sci. 74:2067–2075.[Abstract]

Herpin, P., F. Wosiak, J. Le Dividich, and R. Bertin. 1998. Effects of acute asphyxia at birth on subsequent heat production capacity in newborn pigs. Res. Vet. Sci. 66:45–49.

Jeffreys, H. 1984. Theory of Probability. 3rd ed. Clarendon Press, Oxford, U.K.

Kass, R. E., and A. E. Raftery. 1995. Bayes factors. J. Am. Stat. Assoc. 90:773–795.

Kerr, J. C., and R. Cameron. 1995. Reproductive performance of pigs selected for components of efficient lean growth. Anim. Sci. 60:281–290.

Knol, E. F., J. I. Leenhouwers, and T. van der Lende. 2002. Genetic aspects of piglet survival. Livest. Prod. Sci. 78:47–55.

Lacy, R. C., and C. B. Lynch. 1978. Genetic analysis of characters contributing to temperature regulation in Mus musculus. Part I. Covariance of relatives. Behav. Genet. 8:104–108.

Lynch, C. B., D. S. Sulzbach, and M. S. Connolly. 1988. Quantitative genetic analysis of temperature regulation in Mus domesticus. IV. Pleiotropy and genotype-by-environment interaction. Am. Nat. 132:521–537.

Nespolo, R. F., L. D. Bacigalupe, and F. Bozinovic. 2003. Heritability of energetics in a wild mammal, the leaf-eared mouse (Phyllotis darwini). Evolution 57:1679–1688.[Medline]

Raftery, A. E., and S. M. Lewis. 1996. Implementing MCMC. Pages 115–130 in Markov Chain Monte Carlo in Practice. W. R. Gilks, S. Richardson, and D. J. Spiegelhalter, ed. Chapman & Hall, London, U.K.

Randall, G. C. B. 1971. The relationship of arterial blood pH and pCO₂ to the viability of the newborn piglet. Can. J. Comp. Med. 35:141–146.[Medline]

Roehe, R. 1999. Genetic determination of individual birth weight and its association with sow productivity traits using Bayesian analyses. J. Anim. Sci. 77:330–343.[Abstract/Free Full Text]

Singh, J. P., M. G. Larson, C. J. O’Donnell, H. Tsuji, J. C. Evans, and D. Levy. 1999. Heritability of heart rate variability. The Framingham heart study. Circulation 99:2251–2254.[Abstract/Free Full Text]

Silcock, M., and P. A. Parsons. 1973. Temperature preference differences between strains of Mus musculus, associated variables, and ecological implications. Oecologia 12:147–160.

Sugiyama, F., G. A. Churchill, R. Li, L. J. M. Libby, T. Carver, K. Yagami, S. W. M. John, and B. Paigen. 2002. QTL associated with blood pressure, heart rate, and heart weight in CBA/CaJ and BALB/cJ mice. Physiol. Genomics 10:5–12.[Abstract/Free Full Text]

Tuchscherer, M., B. Puppe, A. Tuchscherer, and U. Tiemann. 2000. Early identification of neonates at risk: traits of newborn piglets with respect to survival. Theriogenology 54:371–388.[Medline]

Varona, L., L. A. García-Cortés, and M. Pérez-Enciso. 2001. Bayes factors for detection of quantitative trait loci. Genet. Sel. Evol. 33:133–152.[Medline]

This article has been cited by other articles:

				X. F. de Sevilla, E. Fabrega, J. Tibau, and J. Casellas Genetic background and phenotypic characterization over two farrowings of leg conformation defects in Landrace and Large White sows J Anim Sci, May 1, 2009; 87(5): 1606 - 1612. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Varona, L. Articles by Noguera, J. L. Search for Related Content PubMed PubMed Citation Articles by Varona, L. Articles by Noguera, J. L.