grapes

(Greetings_here_is_yet_another) Rate of Adaptive Protein Evolution Software

Introduction

grapes is a program estimating the of adaptive and non-adaptive amino-acid substitution rate from polymorphism and divergence coding sequence data. It is written in C++ and based on the Bio++ libraries. This file is relevant to version 1.0. grapes essentially re-implements Adam Eyre-Walker's DoFE program (Eyre-Walker and Keightley 2009) using a different optimization procedure and with a couple of additional features. The details of the method are available from Galtier (2016). A similar approach and program have been developed by Tataru et al. (2017). See Rousselle et al. (2018) for a comparison of the methods and clarification of their relationships.

multigrapes is a program fitting a population genetic model to synonymous + non-synonymous Site Frequency Spectrum (SFS) data, and is essentially a multi-SFS version of the program grapes. This file is relevant to version 1.1. The main new feature of multi_grapes, compared to grapes, is the option of fitting model parameters assumed to be shared by distinct data sets – specifically, the shape of the distribution of fitness effects (DFE). This is the main focus of Galtier & Rousselle 2020, and the reason why this program was developed. In addition, new models of the DFE are implemented in multi_grapes that are absent in grapes. These new option come with costs: (1) several of the models implemented in grapes are not available in multi_grapes; (2) no estimate of the adaptive substitution rate is performed by multi_grapes.

Installation

The distributed binaries for GNU/Linux are statically linked and ready to run.

Compilation

If, for any reason, a source compilation is needed, here are the dependencies :

• gcc >= 4.4
• libgsl >= 1.0
• Bio++. The source code is ready for the latest development version of the libraries. Stable version >= 2.0 (bpp-core, bpp-phyl, bpp-seq) can be used with little modification of the code.

Compilation can be achieved using cmake:

mkdir build
cd build
cmake -DCMAKE_INSTALL_PREFIX=$HOME/.local ..
make make

See the INSTALL file and cmake documentation for more detailed instructions.

Execution

grapes -in input_file.dofe -out output_file.csv -model model_name [options]

multigrapes -in input_file.dofe -out output_file.csv -model model_name [options]

Input

grapes needs information on the number of synonymous and non-synonymous substitutions between focal and outgroup species, number and frequency of synonymous and non-synonymous SNPs (SFS),number of synonymous and non-synonymous sites in the divergence and polymorphism data sets. These are passed via a DoFE file as described in the documentation of the DoFE program: [http://www.lifesci.susx.ac.uk/home/Adam_Eyre-Walker/Website/Software.html][1]. grapes can handle both folded (as in DoFE) and unfolded SFS data. If SFS's are unfolded, this must be indicated via an extra line in the input file containing: #unfolded . See examples at the bottom of this file.

multi_grapes needs information on the number and frequency of synonymous and non-synonymous SNPs (SFS) and number of synonymous and non-synonymous sites, for one or several data sets. These are passed via a DoFE file as described in the documentation of the DoFE program: [http://www.lifesci.susx.ac.uk/home/Adam_Eyre-Walker/Website/Software.html][1]. multi_grapes can handle both folded (as in DoFE) and unfolded SFS data. If SFS's are unfolded, this must be indicated via an extra line in the input file containing: #unfolded. Multiple data sets should appear as different lines of the DoFE file (see examples at the bottom of this file)

Analysis

grapes will estimate the distribution of fitness effect of mutations (DFE), rate of adaptive evolution (), rate of non-adaptive evolution (), and proportion of adaptive substitutions (a) by fitting a population genetic model to SFS + divergence data in the maximum likelihood framework. grapes will also perform more basic analyses, namely estimating a as , referred to as Neutral model, and using the corrected version of Fay, Wickoff and Wu (2002 Nature 415:1024), referred to as FWW.

multi_grapes will estimate the distribution of fitness effect of mutations (DFE) by fitting a population genetic model to SFS + divergence data in the maximum likelihood framework. Data sets will first be analyzed separately (= with all parameters specific to each data set, as in grapes) then jointly (= with one parameter, the shape of the DFE, shared among data sets).

Output

Basic output is written in the terminal. This corresponds to estimates of the adaptive and non-adaptive rates, main model parameters and likelihoods. Detailed output is written in a csv file.

Grapes options

The most important option is the name of the assumed DFE model. Five distinct models are implemented in addition to the Neutral model, namely GammaZero (=Gamma), GammaExpo, DisplGamma, ScaledBeta, and FGMBesselK. See Galtier (2016) for details on what these models mean. One can either use one of the six models (e.g., -model GammaExpo), or do the six in a single run by passing -model all. GammaZero with all default options is equivalent to the second approach of Eyre-Walker and Keightley (2009), with nuisance parameters 's.

Additional options:

*-nearly_neutral <float>: defines the threshold of $Ne.s$ above which a mutation is considered adaptive ($S_{adv}$ in Galtier 2016, default=5) *-FWW_threshold <float>: minimal allele frequency in FWW alpha estimation (default=0.15) *-no_div_param: implements so-called [-A] version in Galtier (2016); also called a DFE by Tataru et al. (2017), and indicated with * in Rousselle et al. (2018); irrelevant if model = GammaZero; (default=false) *-no_syn_orient_error: force equal synonymous and non-synonymous mis-orientation rate(default=false) *-nb_rand_start <int>: number of random starting values in model optimization (default=0); setting positive values will slow down the program but decrease the probability of being trapped in local optima. *-fixed_param <control_file_name>: this option should be used if one does not want to optimize every parameter, but rather have some parameters fixed to predefined values; parameter names and predefined values are passed via a control file; see example at the bottom of this file (default=none).

Multigrapes options

The most important option is the name of the assumed DFE model. Two distinct models are implemented in addition to the Neutral model, namely GammaZero (=Gamma) and ReflectedGamma. See Galtier (2016) and Galtier & Rousselle (2020) for details on what these models mean. In addition, one can add a class of lethal mutations using the -p_lethal option (see below).

Additional options:

• -fold : fold the SFS (if unfolded)
• -no_syn_orient_error : force equal synonymous and non-synonymous mis-orientation rate (default=false)
• -nb_rand_start <int> : number of random starting values in model optimization (default=0); setting positive values will slow down the program but decrease the probability of being trapped in local optima.
• -no_separate : do not perform separate analysis.
• -no_shared : do not perform shared analysis.
• -shared_shape <int>,<int>,...: fixed shared shape parameters (all the listed values will be fitted successively); in the absence of this option, the shared shape parameter will be optimized.
• -p_lethal <float> : proportion of lethal mutations (this parameter cannot be optimized).

Developed by

Nicolas Galtier Institute of Evolutionary Sciences CNRS – University Montpellier

Supported by

Montpellier Bioinformatics & Biodiversity platform [http://mbb.univ-montp2.fr]

References

• Eyre-Walker A, Keightley PD. 2009 Estimating the rate of adaptive molecular evolution in the presence of slightly deleterious mutations and population size change. Molecular Biology and Evolution. 26:2097–2108.
• Galtier N. 2016. Adaptive protein evolution in animals and the effective population size hypothesis. PLoS Genetics 12:e1005774.
• Rousselle M., Mollion M., Nabholz B., Bataillon T., Galtier N. 2018. Overestimation of the adaptive substitution rate in fluctuating populations. Biology Letters (in press)
• Tataru P, Mollion M, Glémin S, Bataillon T. 2017 Inference of distribution of fitness effects and proportion of adaptive substitutions from polymorphism data. Genetics 207:1103–1119.
• Galtier N., Rousselle M. 2020. How much does Ne vary among species? biorxiv [https://doi.org/10.1101/861849]

Example input files for Grapes:

folded DoFE file:

Allolobophora_chlorotica+Aporerctodea_icterica (4852 genes)
all_genes 12 254286 4360.47 1987.61 1339.62 1640.54 1087.89 581.969 92842.3 18049.3 9568.58 6722.38 6820.24 5353.69 2607.35 340000 10578 111000 24890

First line is a header/comment line. Second line contains the data:

• entry 1 ( all_genes ): any string (dataset description)
• entry 2 ( 12 ): sample size (here, 6 diploid individuals)
• entry 3 ( 254286 ): number of non-synonymous sites, polymorphism data
• entry 4 → 3+n/2, where n equals sample size: non-synonymous SFS (number of non-synonymous singletons, doubletons, etc...)
• entry 3+n/2+1 ( 92842.3 ): number of synonymous sites, polymorphism data
• entry 3+n/2+2 → 3+n/2+1+n/2, where n equals sample size: synonymous SFS (number of synonymous singletons, doubletons, etc...)
• entry 3+n/2+1+n/2+1 ( 340000 ): number of non-synonymous sites, divergence data
• entry 3+n/2+1+n/2+2 ( 10578 ): number of non-synonymous substitutions, divergence data
• entry 3+n/2+1+n/2+3 ( 111000 ): number of synonymous sites, divergence data
• entry 3+n/2+1+n/2+4 ( 24890 ): number of synonymous substitutions, divergence data

unfolded DoFE file:

Microtus_arvalis+Microtus_glareolus.fas (2943 genes)
#unfolded
all_genes 12 1.61188e+06 2008.19 590.033 165.33 109.824 85.956 73.6923 61.4066 63.5934 65.3077 104.637 119.604 361114 5046.4 2095.86 836.505 533.066 403.44 357.615 327.066 318.044 322.549 518.198 757.527 1.48228e+06 13089 432574 38940

Same as above with additional #unfolded line, and SFS's containing n-1 entries instead of n/2.

Example control file:

GammaExpo,negGshape,0.4
GammaExpo,negGmean,10000

Each line contains model name, parameter name and fixed parameter value. Parameter names are listed in the table below:

Model	Parameter name	Parameter meaning
GammaZero	negGmean	Gamma distribution mean, negative effects
GammaZero	negGshape	Gamma distribution shape, negative effects
GammaExpo	negGmean	Gamma distribution mean, negative effects
GammaExpo	negGshape	Gamma distribution shape, negative effects
GammaExpo	posGmean	Exponential distribution mean, positive effects
GammaExpo	pos_prop	Proportion of beneficial mutations (positive s)
DisplGamma	negGmean	Un-displaced Gamma distribution mean
DisplGamma	negGshape	Un-displaced Gamma distribution shape
DisplGamma	s0	Displacement
ScaledBeta	Ba	Beta distribution first shape parameter
ScaledBeta	Bb	Beta distribution second shape parameter
ScaledBeta	med_prop	Proportion of mutations with Ne.s above -25
FGMBesselK	BKm	m in equation 8, Lourenco et al 2011*
FGMBesselK	BKnsz2	n/z 2 in equation 8, Lourenco et al 2011*
FGMBesselK	BKsigma	s in equation 8, Lourenco et al 2011*
FGMBesselK	BKscale	scaling parameter, FGM Bessel distribution

*Lourenco et al. 2011 Genetics 65:1559