Limits...
A framework for inferring fitness landscapes of patient-derived viruses using quasispecies theory.

Seifert D, Di Giallonardo F, Metzner KJ, Günthard HF, Beerenwinkel N - Genetics (2014)

Bottom Line: The sampler can overcome situations where no maximum-likelihood estimator exists, and it can adaptively learn the posterior distribution of highly correlated fitness landscapes without prior knowledge of their shape.We tested our approach on simulated data and applied it to clinical human immunodeficiency virus 1 samples to estimate their fitness landscapes in vivo.The posterior fitness distributions allowed for differentiating viral haplotypes from each other, for determining neutral haplotype networks, in which no haplotype is more or less credibly fit than any other, and for detecting epistasis in fitness landscapes.

View Article: PubMed Central - PubMed

Affiliation: Department of Biosystems Science and Engineering, ETH Zurich, Basel 4058, Switzerland Swiss Institute of Bioinformatics, Basel 4058, Switzerland.

Show MeSH

Related in: MedlinePlus

Rank correlation coefficient τKendall for different L and K. The rows depict results for increasingly high-dimensional DNA spaces, where n = 4L denotes the number of haplotypes. The columns depict the density estimators for the rank correlation coefficients between an estimator and the true fitness landscape with increasing K. Densities with dark shading represent τKendall for the QuasiFit-based estimator and densities with light shading represent τKendall for the count-based estimator.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4286684&req=5

fig6: Rank correlation coefficient τKendall for different L and K. The rows depict results for increasingly high-dimensional DNA spaces, where n = 4L denotes the number of haplotypes. The columns depict the density estimators for the rank correlation coefficients between an estimator and the true fitness landscape with increasing K. Densities with dark shading represent τKendall for the QuasiFit-based estimator and densities with light shading represent τKendall for the count-based estimator.

Mentions: To compare our model-based predictions to those of merely using the ranks of the estimated frequencies as a proxy for the ranks of the fitness landscape, we used Kendall’s τ as a measure of agreement in the ranks of different methods of estimating fitness landscapes (Figure 6). The case K = 0 represents fitness landscapes possessing only main/additive effects; i.e., there is no epistasis and as such we can envision a Mount Fuji-like fitness landscape, where mutations will cause the population to ultimately climb to the maximum fitness, as there is only one local optimum that is also the global optimum. The ranks of the fitness landscape and its equilibrium population distribution closely agree in this case (Figure 6). For K > 0, our fitness landscape estimates recover the ranks of the true fitness landscape significantly better than the naive count-based estimator as our model accounts for mutational neighborhood structure (Figure 6).


A framework for inferring fitness landscapes of patient-derived viruses using quasispecies theory.

Seifert D, Di Giallonardo F, Metzner KJ, Günthard HF, Beerenwinkel N - Genetics (2014)

Rank correlation coefficient τKendall for different L and K. The rows depict results for increasingly high-dimensional DNA spaces, where n = 4L denotes the number of haplotypes. The columns depict the density estimators for the rank correlation coefficients between an estimator and the true fitness landscape with increasing K. Densities with dark shading represent τKendall for the QuasiFit-based estimator and densities with light shading represent τKendall for the count-based estimator.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4286684&req=5

fig6: Rank correlation coefficient τKendall for different L and K. The rows depict results for increasingly high-dimensional DNA spaces, where n = 4L denotes the number of haplotypes. The columns depict the density estimators for the rank correlation coefficients between an estimator and the true fitness landscape with increasing K. Densities with dark shading represent τKendall for the QuasiFit-based estimator and densities with light shading represent τKendall for the count-based estimator.
Mentions: To compare our model-based predictions to those of merely using the ranks of the estimated frequencies as a proxy for the ranks of the fitness landscape, we used Kendall’s τ as a measure of agreement in the ranks of different methods of estimating fitness landscapes (Figure 6). The case K = 0 represents fitness landscapes possessing only main/additive effects; i.e., there is no epistasis and as such we can envision a Mount Fuji-like fitness landscape, where mutations will cause the population to ultimately climb to the maximum fitness, as there is only one local optimum that is also the global optimum. The ranks of the fitness landscape and its equilibrium population distribution closely agree in this case (Figure 6). For K > 0, our fitness landscape estimates recover the ranks of the true fitness landscape significantly better than the naive count-based estimator as our model accounts for mutational neighborhood structure (Figure 6).

Bottom Line: The sampler can overcome situations where no maximum-likelihood estimator exists, and it can adaptively learn the posterior distribution of highly correlated fitness landscapes without prior knowledge of their shape.We tested our approach on simulated data and applied it to clinical human immunodeficiency virus 1 samples to estimate their fitness landscapes in vivo.The posterior fitness distributions allowed for differentiating viral haplotypes from each other, for determining neutral haplotype networks, in which no haplotype is more or less credibly fit than any other, and for detecting epistasis in fitness landscapes.

View Article: PubMed Central - PubMed

Affiliation: Department of Biosystems Science and Engineering, ETH Zurich, Basel 4058, Switzerland Swiss Institute of Bioinformatics, Basel 4058, Switzerland.

Show MeSH
Related in: MedlinePlus