Academic journal article Genetics

Pleiotropy Can Be Effectively Estimated without Counting Phenotypes through the Rank of a Genotype-Phenotype Map

Academic journal article Genetics

Pleiotropy Can Be Effectively Estimated without Counting Phenotypes through the Rank of a Genotype-Phenotype Map

Article excerpt

(ProQuest: ... denotes formulae omitted.)

PLEIOTROPY, or the capability of a gene to affect multiple phenotypes (Fisher 1930; Wright 1968), has played a cen- tral role in genetics, development, and evolution (e.g.,Lande 1980; Turelli 1985; Wagner 1989; Barton 1990; Keightley 1994; Hartl and Taubes 1996, 1998; Waxman and Peck 1998; Lynch et al. 1999; Bataillon 2000; Orr 2000; Poon and Otto 2000; Wagner 2000; Elena and Lenski 2003; Welch and Waxman 2003; Wingreen et al. 2003; Zhang and Hill 2003; MacLean et al. 2004; Otto 2004; Eyre-Walker et al. 2006; Pigliucci 2008; Wagner et al. 2008). Although func- tional genomics has brought high-throughput data to bear on the nature and extent of pleiotropy (Dudley et al. 2005; Ohya et al. 2005; Pal et al. 2006; Cooper et al. 2007), this issue remains highly controversial, largely because of the phenotypic complexity; see Wagner and Zhang (2011), Hill and Zhang (2012), and Paaby and Rockman (2013) for recent reviews and comments.

Meanwhile, a new research line has emerged in the past decade, aiming at the estimation of gene pleiotropy from genetics or sequence data, rather than the measurement of affected phenotypes (Martin and Lenormand 2006; Gu 2007a, b; Chevin et al. 2010; Su et al. 2010; Zeng and Gu 2010; Razeto-Barry et al. 2012; Chen et al. 2013). For instance, Gu (2007a) developed a statistical method to estimate the "effec- tive pleiotropy" (Ke) of a gene from the multiple-sequence align- ment (MSA) of protein sequences. Most genes have Ke in the range between 1 and 20 (Su et al. 2010), with the medium Ke = 6.5 of these estimates that is comparable to some empir- ical pleiotropy measures (Wagner and Zhang 2011; Paaby and Rockman 2013). However, we have to acknowledge that this approach is not very easy to be understood; that is, How can we know the amount of multifunctionality (pleiotropy) of a gene without biologically knowing each functional component?

In this article we try to answer the following questions: What quantity was actually estimated by the method proposed by Gu (2007a)? And under which condition can this estimate be interpreted as effective gene pleiotropy? To answer these questions, we introduce the concept of minimum pleiotropy (Pmin), which has an intrinsic relationship with the rank of the genotype-phenotype map. Therefore, we propose that the method of Gu (2007a) is to estimate the rank of the genotype- phenotype map that can be regarded as effective estimation of Pmin. Simulations and empirical studies are carried out to test several predictions from the new theory.

Materials and Methods

Data sets

Random single-copy gene samples were from three sets of model organisms, respectively. MSAs of vertebrate genes were from the data set of Su et al. (2010), while those of five yeast homologous genes (Saccharomyces cerevisiae, Candida glabrata, Kluyveromyces lactis, Debaryomyces hansenii, and Yarrowia lipolytica) were from the Génolevures database (, and those of 12 Drosophila genes were from the Consortium of Drosophila genome projects.

Estimation of Ke from protein sequences

We calculated the dN/dS (the ratio of nonsynonymous to synonymous rates) of each gene, based on the orthologous sequences from two closely related species: human and mouse for the vertebrate data set, S. cerevisiae and S. para- doxus for the yeast data set, and D. melanogaster and D. sechellia for the Drosophila data set. After the gene phylogeny was reconstructed by the neighbor-joining method, we esti- mated H, a normalized index for the rate variation among sites (Gu 2007a). It follows that the effective gene pleiotropy (Ke) was estimated by solving the equation [dN/dS]/(1 2 H)= 22Kj[1 +u(Kj)], where u(Ke)=0.0208Ke(Ke +2)/(1+ 0.289Ke). The technical procedures for estimating Ke are de- scribed by Gu (2007a) and Su et al. (2010) in detail.

Estimation of Ke from nucleotide sequences

Vertebrate genes used in this analysis were from the data set of Su et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.