 Software
 Open Access
Software for the analysis and visualization of deep mutational scanning data
 Jesse D Bloom^{1}Email author
https://doi.org/10.1186/s1285901505904
© Bloom; licensee BioMed Central. 2015
 Received: 9 January 2015
 Accepted: 22 April 2015
 Published: 20 May 2015
Abstract
Background
Deep mutational scanning is a technique to estimate the impacts of mutations on a gene by using deep sequencing to count mutations in a library of variants before and after imposing a functional selection. The impacts of mutations must be inferred from changes in their counts after selection.
Results
I describe a software package, dms_tools, to infer the impacts of mutations from deep mutational scanning data using a likelihoodbased treatment of the mutation counts. I show that dms_tools yields more accurate inferences on simulated data than simply calculating ratios of counts pre and postselection. Using dms_tools, one can infer the preference of each site for each amino acid given a single selection pressure, or assess the extent to which these preferences change under different selection pressures. The preferences and their changes can be intuitively visualized with sequencelogostyle plots created using an extension to weblogo.
Conclusions
dms_tools implements a statistically principled approach for the analysis and subsequent visualization of deep mutational scanning data.
Keywords
 Deep mutational scanning
 Sequence logo
 Aminoacid preferences
Background
A key component of deep mutational scanning is analysis of the data: First, raw reads from the deep sequencing must be processed to count mutations pre and postselection. Next, the biological effects of mutations must be inferred from these counts. The first task of processing the reads is idiosyncratic to the specific sequencing strategy used. But the second task of inferring mutational effects from sequencing counts is amenable to more general algorithms. However, only a few such algorithms have been described [19,20]. Here I present userfriendly software, dms_tools, that infers mutational effects from sequencing counts. Before describing the algorithms implemented in dms_tools and illustrating its use on existing and simulated data, I first discuss issues associated with inferring mutational effects from sequencing counts.
The nature of deep mutational scanning data.
The data consist of counts of variants pre and postselection. The approach presented here treats each site in the gene separately, ignoring epistatic coupling among mutations. This aspect of the approach should not be construed as a suggestion that interactions among mutations are unimportant; indeed, several studies have used deep mutational scanning to examine pairwise epistasis [14,21,22], and techniques have been described to obtain linkage between distant sites [23,24]. However, the exploding combinatorics of multiple mutations (a 500residue protein has only 19×500≈10^{4} single mutants, but \(19^{2} \times \frac {500 \times 499}{2} \approx 4 \times 10^{7}\) double mutants and \(19^{3} \times \frac {500!}{497! \times 3!} \approx 10^{11}\) triple mutants) make it currently plausible to comprehensively characterize only single mutations to all but the shortest genes. Treating sites independently is therefore not a major limitation for most current datasets. Eventually the approach here might be extended to include coupling among mutations.
The data for each site r is characterized by the sequencing depth (total number of counts); let \(N_{r}^{\text {pre}}\), \(N_{r}^{\text {post}}\), \(N_{r}^{s1}\), and \(N_{r}^{s2}\) denote the depth at r for each of the four libraries in Figure 1 (preselection, postselection, selection s1, and selection s2). Typical depths for current experiments are N∼10^{6}. Denote the counts of character x (characters might be nucleotides, amino acids, or codons) at r as \(n_{r,x}^{\text {pre}}\), \(n_{r,x}^{\text {post}}\), \(n_{r,x}^{s1}\), and \(n_{r,x}^{s2}\). The values of n _{ r,x } for characters x that differ from the wildtype identity wt(r) depend on both the depth N and the average persite mutation rate \(\overline {\mu }\). Since the mutations are intentionally introduced into the mutant library by the experimentalist, in principle \(\overline {\mu }\) could have any value. But typically, deep mutational scanning experiments aim to introduce about one mutation per gene to avoid filling the mutant library with highly mutated genes – so the average mutation rate is usually \(\overline {\mu } \sim 1/L\) where L is the length of the gene. Therefore, if a 500codon gene is sequenced at depth N∼10^{6}, we expect \(N \overline {\mu } \sim 2000\) counts of nonwildtype codons at each site. Since there are 63 mutant codons, the average preselection counts for a mutation to a specific x≠ wt(r) will be \(n_{r,x}^{\text {pre}} \sim 30\), with counts for most mutations deviating from this average due to biases in creation of the mutant library and randomness in which molecules are sequenced. Counts in the postselection libraries will further deviate from this average due to selection. Therefore, even at depths N∼10^{6}, the actual counts of most mutations will be quite modest.
The rest of this paper assumes that the sequencing depth is less than the number of unique molecules in the mutant library, such that the deep sequencing randomly subsamples the set of molecules. If this assumption is false (i.e. if the number of unique molecules is substantially less than the sequencing depth), then the accuracy of inferences about mutational effects will be fundamentally limited by this aspect of the experimental design. Properly done experiments should quantify the number of unique molecules in the library so that it is obvious whether this assumption holds. In the absence of such information, the analysis can be repeated using only a random fraction of the deep sequencing data to assess whether inferences are limited by sequencing depth or the underlying molecular diversity in the mutant library.
The goal: inferring sitespecific aminoacid preferences
The goal is to estimate the effects of mutations from changes in their counts after selection. Let μ _{ r,x }, f _{ r,x }, \(f_{r,x}^{s1}\), and \(f_{r,x}^{s2}\) denote the true frequencies at site r of all mutant characters x≠ wt(r) that would be observed for the four libraries in Figure 1 if we sampled at infinite depth in both the actual experiment and the sequencing. The definition of these frequencies for the wildtype character wt(r) depends on how the mutant library is constructed. If the mutant library is constructed so that there is a Poisson distribution of the number of mutations per gene (as is the case for errorprone PCR or the codonmutagenesis in [9,11]), then μ _{ r,wt(r)}, f _{ r,wt(r)}, \(f_{r,\operatorname {wt}\left (r\right)}^{s1}\), and \(f_{r,\operatorname {wt}\left (r\right)}^{s2}\) are defined as for all other characters x, and denote the frequencies of wt(r) at site r that would be observed if sampling at infinite depth. The reason we can make this definition for libraries containing genes with Poissondistributed numbers of mutations is that for any reasonablelength gene (L≫1), the marginal distribution of the number of mutations in a gene is virtually unchanged by the knowledge that there is a mutation at site r. On the other hand, if the mutant library is constructed so that there is exactly zero or one mutation per gene (as in [8,10,15]), then the marginal distribution of the total number of mutations in a gene is changed by the knowledge that there is a mutation at r. In this case, the wildtypecharacter frequencies μ _{ r,wt(r)}, f _{ r,wt(r)}, \(f_{r,\operatorname {wt}\left (r\right)}^{s1}\), and \(f_{r,\operatorname {wt}\left (r\right)}^{s2}\) are correctly defined as the frequency of unmutated genes in the library, and the counts \(n_{r,\operatorname {wt}\left (r\right)}^{\text {pre}}\), etc. are defined as the number of reads at r attributable to unmutated genes. In this case, measurement of these counts requires sequencing with linkage as in [15,23,24]. The proper analysis of libraries containing only unmutated and singly mutated clones sequenced without linkage is beyond the scope of this paper.
where y is summed over all character identities (all nucleotides, codons, or amino acids). The preferences can be intuitively visualized (Figure 1A) and interpreted as the equilibrium frequencies in substitution models for gene evolution [9,25] (after accounting for uneven mutational rates [26,27]).
The challenge of statistical inference from finite counts
Equations 1 and 2 are in terms of the true frequencies μ _{ r,x }, f _{ r,x }, etc. But in practice, we only observe the counts in the finite sample of sequenced molecules. The computational challenge is to estimate the preferences (or enrichment ratios) from these counts.
where (often chosen to be one) is a pseudocount added to each count to avoid ratios of zero or infinity.
However, Equation 4 involves ratios of counts with values ∼10 to 100 – and as originally noted by Karl Pearson [28,29], ratios estimated from finite counts are statistically biased, with the bias increasing as the magnitude of the counts decrease. This bias can propagate into subsequent analyses, since many statistical tests assume symmetric errors. The problems caused by biases in uncorrected ratios have been noted even in applications such as isotoperatio mass spectrometry [30] and fluorescent imaging [31], where the counts usually far exceed those in deep mutational scanning.
Taking ratios also abrogates our ability to use the magnitude of the counts to assess our certainty about conclusions. For instance, imagine that at a fixed depth, the counts of a mutation increase from a preselection value of 5 to a postselection value of 10. While this doubling suggests that the mutation might be beneficial, the small counts make us somewhat uncertain of this conclusion. But if the counts increased from 20 to 40 we would be substantially more certain, and if they increased from 100 to 200 we would be quite sure. So only by an explicit statistical treatment of the counts can we fully leverage the data.
Here I describe a software package, dms_tools, that infers mutational effects in a Bayesian framework using a likelihoodbased treatment of the counts. This software can be used to infer and visualize sitespecific preferences from experiments like Figure 1A, and to infer and visualize differences in preferences under alternative selections from experiments like Figure 1B.
Implementation and results
Algorithm to infer sitespecific preferences
dms_tools uses a Bayesian approach to infer sitespecific preferences from experiments like those in Figure 1A. The algorithm calculates the likelihoods of the counts given the unknown preferences and mutation/error rates, placing plausible priors over these unknown parameters. The priors correspond to the assumption that all possible identities (e.g. amino acids) have equal preferences, and that the mutation and error rates for each site are equal to the overall average for the gene. MCMC is used to calculate the posterior probability of the preferences given the counts.
This algorithm is a slight modification of that in the Methods of [9]; here the algorithm is described anew to explain the implementation in dms_tools.
Optional controls to quantify error rates
Some sequencing reads that report a mutation may actually reflect an error introduced during sequencing or PCR rather than an actual mutation that experienced selection. Errors can be quantified by sequencing an unmutated gene, so that any counts at r of x≠ wt(r) for this control reflect errors. In some cases (e.g. sequencing an RNA virus where the postselection libraries must be reversetranscribed), error rates for the pre and postselection libraries may differ and so be described by different controls. Let \(N_{r}^{\text {errpre}}\) and \(N_{r}^{\text {errpost}}\) be the depth and \(n_{r,x}^{\text {errpre}}\) and \(n_{r,x}^{\text {errpost}}\) be the counts of x in the preselection and postselection error controls, respectively. Define ε _{ r,x } and ρ _{ r,x } to be the true frequencies of errors at r from wt(r) to x in the pre and postselection controls, respectively.
Likelihoods of observing specific mutational counts
Define vectors of the counts and frequencies for all characters at each site r, i.e. \(\mathbf {n_{r}^{\textbf {pre}}}= \left (\cdots, n_{r,x}^{\text {pre}}, \cdots \right)\), \(\mathbf {n_{r}^{\textbf {post}}}= \left (\cdots, n_{r,x}^{\text {post}}, \cdots \right)\), μ _{ r }=(⋯,μ _{ r,x },⋯), f _{ r }=(⋯,f _{ r,x },⋯), etc. Also define π _{ r }=(⋯,π _{ r,x },⋯) of the preferences for each r, noting that Equation 3 implies \(\boldsymbol {\mathbf {f_{r}}} = \frac {\boldsymbol {\mathbf {\mu _{r}}}\circ \boldsymbol {\mathbf {\pi _{r}}}}{\boldsymbol {\mathbf {\mu _{r}}}\cdot \boldsymbol {\mathbf {\pi _{r}}}}\) where ∘ is the Hadamard product.
where Multi is the multinomial distribution, δ _{ r }=(⋯,δ _{ x,wt(r)},⋯) is a vector with the element corresponding to wt(r) equal to one and all other elements zero (δ _{ xy } is the Kronecker delta), and we have assumed that the probability that a site experiences both a mutation and an error is negligibly small.
Priors over the unknown parameters
where \(\mathcal {C}_{m}\) is the number of mutant characters with m changes relative to wildtype (for nucleotides \(\mathcal {C}_{0} = 1\) and \(\mathcal {C}_{1} = 3\); for codons \(\mathcal {C}_{0} = 1\), \(\mathcal {C}_{1} = 9\), \(\mathcal {C}_{2} = \mathcal {C}_{3} = 27\)).
Character types: nucleotides, amino acids, or codons
dms_tools allows four possibilities for the type of character for the counts and preferences. The first three possibilities are simple: the counts and preferences can both be for any of nucleotides, amino acids, or codons.
where w is a 64element vector of codons x, A(w) is a 20 or 21element (depending on the treatment of stop codons) vector of amino acids a, and \(\mathcal {A}\left (x\right)\) is the amino acid encoded by x. The prior over the preferences π _{ r } is still a symmetric Dirichlet (now only of length 20 or 21), but the priors for μ _{ r }, ε _{ r }, and ρ _{ r } are now Dirichlets parameterized by A(a _{ r,μ }), A(a _{ r,ε }) and A(a _{ r,ρ }) rather than a _{ r,μ }, a _{ r,ε }, and a _{ r,ρ }. The likelihoods are computed in terms of these transformed vectors after similarly transforming the counts to \(\mathbf {A}\left (\mathbf {n_{r}^{\textbf {pre}}}\right)\), \(\mathbf {A}\left (\mathbf {n_{r}^{\textbf {post}}}\right)\), \(\mathbf {A}\left (\mathbf {n_{r}^{\textbf {errpre}}}\right)\), and \(\mathbf {A}\left (\mathbf {n_{r}^{\textbf {errpost}}}\right)\).
Implementation
The program dms_inferprefs in the dms_tools package infers the preferences by using pystan [32] to perform MCMC over the posterior defined by the product of the likelihoods and priors in Equations 5, 6, 7, 8, 9, 10, 11, and 12. The program runs four chains from different initial values, and checks for convergence by ensuring that the GelmanRubin statistic \(\hat {R}\) [33] is <1.1 and the effective sample size is >100; the number of MCMC iterations is increased until convergence is achieved. The program dms_logoplot in the dms_tools package visualizes the posterior mean preferences via an extension to weblogo [34]. The program dms_merge can be used to average preferences inferred from different experimental replicates that have individually been analyzed by dms_inferprefs, and the program dms_correlate can be used to compute the correlations among inferences from different replicates.
Inferring preferences with dms_tools
Application to actual datasets
Figures 2 and 3 also illustrate use of dms_correlate to assess the correlation between preferences inferred from different biological replicates [35] of the experiment. The inclusion and analysis of such replicates is the only sure way to fully assess the sources of noise associated with deep mutational scanning.
Testing on simulated data
Is the Bayesian inference worthwhile?
The foregoing sections explain why the Bayesian inference of preferences implemented in dms_tools is conceptually preferable to estimating mutational effects via direct ratio estimation using Equation 4. However, do the practical benefits of this Bayesian inference justify its increased complexity? The simulations in the previous section show that the Bayesian inference is more accurate, but in the absence of background errors (Figure 4A) the magnitude of the improvement becomes negligible once the mutation counts per site \(N \overline {\mu }\) start to exceed ∼10^{3}. When there is a need to correct for background errors (Figure 4B), meaningful benefits of the Bayesian inference over enrichment ratios extend to somewhat higher sequencing depths. Overall, it appears that Bayesian inference will always perform as well or better than ratio estimation, but that the tangible benefit becomes negligible at high sequencing depth. In that case, the user will have to decide if the increased computational runtime and complexity of the Bayesian inference is worth a small improvement in accuracy. Simpler ratio estimation can be performed using the –ratio_estimation option of dms_inferprefs or using an alternative program such as Enrich [19]. When applying ratio estimation to data where some mutations have low counts, it is important to include pseudocounts (denoted by in Equation 4) as a form of regularization to avoid estimating excessively high or low preferences at sites with limited counts.
Algorithm to infer differential preferences
As shown in Figure 1B, a useful extension to the experiment in Figure 1A is to subject the functional variants to two different selection pressures to identify mutations favored by one pressure versus the other. While this experiment could in principle by analyzed by simply comparing the initial unselected mutants to the final variants after the two alternative selections, this approach is nonideal. In experiments like Figure 1A, many mutations are enriched or depleted to some extent by selection, since a large fraction of random mutations affect protein function [3640]. Therefore, the assumption that all mutations are equally tolerated (i.e. the preferences for a site are all equal, or the enrichment ratios are all one) is not a plausible null hypothesis for Figure 1A. For this reason, dms_tools simply infers the preferences given a uniform Dirichlet prior rather than trying to pinpoint some subset of sites with unequal preferences.
But in Figure 1B, the assumption that most mutations will be similarly selected is a plausible null hypothesis, since we expect alternative selections to have markedly different effects on only a small subset of mutations (typically, major constraints related to protein folding and stability will be conserved across different selections on the same protein). Therefore, dms_tools uses a different algorithm to infer the differential preferences under the two selections. This algorithm combines a prior that mildly favors differential preferences of zero with a likelihoodbased analysis of the mutation counts to estimate posterior probabilities over the differential preferences.
Definition of the differential preferences
Likelihoods of observing specific mutational counts
Define vectors of the counts as \(\mathbf {n_{r}^{start}}= \left (\cdots, n_{r,x}^{\text {start}}, \cdots \right)\) for the postselection functional variants that are subjected to the further selections, and as \(\mathbf {n_{r}^{s1}}= \left (\cdots, n_{r,x}^{s1}, \cdots \right)\) and \(\mathbf {n_{r}^{s2}}= \left (\cdots, n_{r,x}^{s2}, \cdots \right)\) for selections s1 and s2. We again allow an error control, but now assume that the same control applies to all three libraries (since they are all sequenced after a selection), and define the counts for this control as \(\mathbf {n_{r}^{\textbf {err}}}= \left (\cdots, n_{r,x}^{\text {err}}, \cdots \right)\); the true error frequencies are denoted by ξ _{ r,x }. Define vectors of the frequencies, errors, control preferences, and differential preferences: \(\mathbf {f_{r}^{\,\textbf {start}}}= \left (\cdots, f_{r,x}^{\text {start}}, \cdots \right)\), \(\mathbf {f_{r}^{\,s1}}= \left (\cdots, f_{r,x}^{s1}, \cdots \right)\), \(\mathbf {f_{r}^{\,s2}}= \left (\cdots, f_{r,x}^{s2}, \cdots \right)\), ξ _{ r }=(⋯,ξ _{ r,x },⋯), \(\boldsymbol {\mathbf {\pi ^{s1}_{r}}}= \left (\cdots, \pi ^{s1}_{r,x}, \cdots \right)\), and Δπ _{ r }=(⋯,Δ π _{ r,x },⋯). Equations 20 and 21 imply \(\mathbf {f_{r}^{\,s1}}= \frac {\mathbf {f_{r}^{\,\textbf {start}}}\circ \mathbf {\pi _{r}^{s1}}}{\mathbf {f_{r}^{\,\textbf {start}}}\cdot \mathbf {\pi _{r}^{s1}}}\) and \(\mathbf {f_{r}^{\,s2}} = \frac {\mathbf {f_{r}^{\textbf {start}}}\circ \left (\mathbf {\pi _{r}^{s1}} + \mathbf {\Delta \pi _{r}}\right)}{\mathbf {f_{r}^{\,\textbf {start}}}\cdot \left (\mathbf {\pi _{r}^{s1}} + \mathbf {\Delta \pi _{r}}\right)}\).
where we have assumed that the probability that a site experiences a mutation and an error in the same molecule is negligibly small.
Priors over the unknown parameters
where dms_tools by default sets all the scalar concentration parameters (α’s) to one except α _{ Δ π }, which is set to two corresponding to a weak expectation that the Δ π values are close to zero. The average error rate \(\overline {\xi _{m}}\) for mutations with m nucleotide changes is
Implementation
The program dms_inferdiffprefs in the dms_ tools package infers the differential preferences by performing MCMC over the posterior defined by the product of the likelihoods and priors in Equations 24, 25, 26, 27, 28, 29, 30, and 31. The MCMC is performed as described for the preferences, and characters can again be any of nucleotides, amino acids, or codons. The program dms_logoplot visualizes the posterior mean differential preferences via an extension to weblogo [34]. In addition, dms_inferdiffprefs creates text files that give the posterior probability that Δ π _{ r,x }>0 or <0. These posterior probabilities are not corrected to account for the fact that multiple sites are typically being examined, although by default the inferences are made using the regularizing prior in Equation 31.
Inferring differential preference with dms_tools
Conclusions
dms_tools is a freely available software package that uses a statistically principled approach to analyze deep mutational scanning data. This paper shows that dms_tools accurately infers preferences and differential preferences from data simulated under realistic parameters. As the figures illustrate, dms_tools can also be applied to actual data with a few simple commands. The intuitive visualizations created by dms_tools assist in interpreting the results. As deep mutational scanning continues to proliferate as an experimental technique [1], dms_tools can be applied to analyze the data for purposes such as guiding protein engineering [3,10], understanding sequencestructurefunction relationships [4,5,7,14,21], informing the development of better evolutionary models for sequence analysis [9,25], and probing the biology of viruses and cells [6,8,1113,18].
Availability and requirements

Project name: dms_tools

Project home page:

Documentation and installation instructions: http://jbloom.github.io/dms_tools/

Source code: https://github.com/jbloom/dms_tools


Operating system(s): Linux

Programming language: Python

Other requirements: pystan, weblogo

License: GNU GPLv3

Restrictions to use by nonacademics: None
Data and code for figures in this paper
The data and computer code used to generate the figures are in version 1.01 of the dms_tools source code (which is tagged on Github at https://github.com/jbloom/dms_tools/tree/1.0.1) in the examples subdirectory. The LaTex source for this paper is in the paper subdirectory.
Declarations
Acknowledgements
Thanks to Alec Heckert for assistance in testing dms_tools, to Erick Matsen for the excellent suggestion to use pystan for MCMC, to Nicholas Wu for providing the mutational counts data from [13], and to Orr Ashenberg and Hugh Haddox for helpful comments on the manuscript. This work was supported by the NIGMS of the NIH under grant R01GM102198.
Authors’ Affiliations
References
 Fowler DM, Fields S. Deep mutational scanning: a new style of protein science. Nat Methods. 2014; 11(8):801–7.View ArticlePubMedPubMed CentralGoogle Scholar
 Fowler DM, Araya CL, Fleishman SJ, Kellogg EH, Stephany JJ, Baker D, et al. Highresolution mapping of protein sequencefunction relationships. Nat Methods. 2010; 7(9):741–6.View ArticlePubMedPubMed CentralGoogle Scholar
 Traxlmayr MW, Hasenhindl C, Hackl M, Stadlmayr G, Rybka JD, Borth N, et al.Construction of a stability landscape of the CH3 domain of human IgG1 by combining directed evolution with high throughput sequencing. J Mol Biol. 2012; 423:397–412.View ArticlePubMedPubMed CentralGoogle Scholar
 McLaughlin Jr RN, Poelwijk FJ, Raman A, Gosal WS, Ranganathan R. The spatial architecture of protein function and adaptation. Nature. 2012; 491(7422):138.View ArticleGoogle Scholar
 Starita LM, Pruneda JN, Lo RS, Fowler DM, Kim HJ, Hiatt JB, et al. Activityenhancing mutations in an E3 ubiquitin ligase identified by highthroughput mutagenesis. Proc Natl Acad Sci USA. 2013; 110(14):1263–72.View ArticleGoogle Scholar
 Melamed D, Young DL, Gamble CE, Miller CR, Fields S. Deep mutational scanning of an RRM domain of the Saccharomyces cerevisiae poly (A)binding protein. RNA. 2013; 19(11):1537–51.View ArticlePubMedPubMed CentralGoogle Scholar
 Roscoe BP, Thayer KM, Zeldovich KB, Fushman D, Bolon DN. Analyses of the effects of all ubiquitin point mutants on yeast growth rate. J Mol Biol. 2013; 425:1363–77.View ArticlePubMedPubMed CentralGoogle Scholar
 Firnberg E, Labonte JW, Gray JJ, Ostermeier M. A comprehensive, highresolution map of a gene’s fitness landscape. Mol Biol Evol. 2014; 31(6):1581–92.View ArticlePubMedPubMed CentralGoogle Scholar
 Bloom JD. An experimentally determined evolutionary model dramatically improves phylogenetic fit. Mol Biol Evol. 2014; 30:1956–78. http://mbe.oxfordjournals.org/content/31/8/1956.View ArticleGoogle Scholar
 Melnikov A, Rogov P, Wang L, Gnirke A, Mikkelsen TS. Comprehensive mutational scanning of a kinase in vivo reveals contextdependent fitness landscapes. Nucleic Acids Res. 2014; 42:112.View ArticleGoogle Scholar
 Thyagarajan B, Bloom JD. The inherent mutational tolerance and antigenic evolvability of influenza hemagglutinin. eLife. 2014; 3:03300. http://elifesciences.org/content/3/e03300.View ArticleGoogle Scholar
 Wu NC, Young AP, AlMawsawi LQ, Olson CA, Feng J, Qi H, et al. Highthroughput profiling of influenza A virus hemagglutinin gene at singlenucleotide resolution. Sci Rep. 2014; 4:4942.PubMedPubMed CentralGoogle Scholar
 Wu NC, Young AP, AlMawsawi LQ, Olson CA, Feng J, Qi H, et al. Highthroughput identification of lossoffunction mutations for antiinterferon activity in the influenza A virus NS segment. J Virol. 2014; 88(17):10157–64.View ArticlePubMedPubMed CentralGoogle Scholar
 Olson CA, Wu NC, Sun R. A comprehensive biophysical description of pairwise epistasis throughout an entire protein domain. Curr Biol. 2014; 24(22):2643–51.View ArticlePubMedPubMed CentralGoogle Scholar
 Kitzman JO, Starita LM, Lo RS, Fields S, Shendure J. Massively parallel singleaminoacid mutagenesis. Nat Methods. 2015; 12:203–6.View ArticlePubMedPubMed CentralGoogle Scholar
 Firnberg E, Ostermeier M. PFunkel: efficient, expansive, userdefined mutagenesis. PLoS One. 2012; 7:52031.View ArticleGoogle Scholar
 Jain PC, Varadarajan R. A rapid, efficient, and economical inverse polymerase chain reactionbased method for generating a site saturation mutant library. Anal Biochem. 2014; 449:90–8.View ArticlePubMedGoogle Scholar
 Findlay GM, Boyle EA, Hause RJ, Klein JC, Shendure J. Saturation editing of genomic regions by multiplex homologydirected repair. Nat. 2014; 513(7516):120–3.View ArticleGoogle Scholar
 Fowler DM, Araya CL, Gerard W, Fields S. Enrich: software for analysis of protein function by enrichment and depletion of variants. Bioinformatics. 2011; 27(24):3430–1.View ArticlePubMedPubMed CentralGoogle Scholar
 Bank C, Hietpas RT, Wong A, Bolon DN, Jensen JD. A bayesian mcmc approach to assess the complete distribution of fitness effects of new mutations: uncovering the potential for adaptive walks in challenging environments. Genet. 2014; 196(3):841–52.View ArticleGoogle Scholar
 Araya CL, Fowler DM, Chen W, Muniez I, Kelly JW, Fields S. A fundamental protein property, thermodynamic stability, revealed solely from largescale measurements of protein function. Proc Natl Acad Sci. 2012; 109(42):16858–63.View ArticlePubMedPubMed CentralGoogle Scholar
 Bank C, Hietpas RT, Jensen JD, Bolon DN. A systematic survey of an intragenic epistatic landscape. Mol Biol Evol. 2015; 32(1):229–38.View ArticlePubMedGoogle Scholar
 Hiatt JB, Patwardhan RP, Turner EH, Lee C, Shendure J. Parallel, tagdirected assembly of locally derived short sequence reads. Nat Methods. 2010; 7(2):119–22.View ArticlePubMedPubMed CentralGoogle Scholar
 Wu NC, De La Cruz J, AlMawsawi LQ, Olson CA, Qi H, Luan HH, et al. HIV1 quasispecies delineation by tag linkage deep sequencing. PloS one. 2014; 9(5):97505.View ArticleGoogle Scholar
 Bloom JD. An experimentally informed evolutionary model improves phylogenetic fit to divergent lactamase homologs. Mol Biol Evol. 2014; 31:2753–769. http://mbe.oxfordjournals.org/content/31/10/2753.View ArticlePubMedPubMed CentralGoogle Scholar
 Yampolsky LY, Stoltzfus A. The exchangeability of amino acids in proteins. Genet. 2005; 170(4):1459–72.View ArticleGoogle Scholar
 Stoltzfus A, Yampolsky LY. Climbing mount probable: mutation as a cause of nonrandomness in evolution. J Hered. 2009; 100(5):637–47.View ArticlePubMedGoogle Scholar
 Pearson K. Mathematical contributions to the theory of evolution. On a form of spurious correlation which may arise when indices are used in the measurement of organs. Proc Royal Society London. 1896; 60(359–367):489–98.View ArticleGoogle Scholar
 Pearson K. On the constants of indexdistributions as deduced from the like constants for the components of the ratio, with special reference to the opsonic index. Biometrika. 1910; 7(4):531–41. doi:10.1093/biomet/7.4.531.View ArticleGoogle Scholar
 Ogliore R, Huss G, Nagashima K. Ratio estimation in SIMS analysis. Nuclear instruments and methods in physics research section B: beam interactions with materials and atoms. 2011; 269(17):1910–18. doi:10.1016/j.nimb.2011.04.120.View ArticleGoogle Scholar
 Van Kempen G, Van Vliet L. Mean and variance of ratio estimators used in fluorescence ratio imaging. Cytometry. 2000; 39(4):300–5.View ArticlePubMedGoogle Scholar
 Stan Development Team. PyStan: the Python interface to Stan, Version 2.5.0. 2014. http://mcstan.org/pystan.html.
 Gelman A, Rubin DB. Inference from iterative simulation using multiple sequences. Stat Sci. 1992; 7:457–72.View ArticleGoogle Scholar
 Crooks GE, Hon G, Chandonia JM, Brenner SE. Weblogo: a sequence logo generator. Genome Res. 2004; 14(6):1188–90. doi:10.1101/gr.849004.View ArticlePubMedPubMed CentralGoogle Scholar
 Blainey P, Krzywinski M, Altman N. Points of significance: replication. Nat Methods. 2014; 11(9):879–80.View ArticlePubMedGoogle Scholar
 Shortle D, Lin B. Genetic analysis of staphylococcal nuclease: identification of three intragenic “global” suppressors of nucleaseminus mutations. Genet. 1985; 110:539–55.Google Scholar
 Rennell D, Bouvier SE, Hardy LW, Poteete AR. Systematic mutation of bacteriophage T4 lysozyme. J Mol Biol. 1991; 222:67–87.View ArticlePubMedGoogle Scholar
 Shafikhani S, Siegel RA, Ferrari E, Schellenberger V. Generation of large libraries of random mutants in Bacillus subtilis by PCRbased plasmid multimerization. Biotechniques. 1997; 23:304–10.PubMedGoogle Scholar
 Guo HH, Choe J, Loeb LA. Protein tolerance to random amino acid change. Proc Natl Acad Sci USA. 2004; 101:9205–210.View ArticlePubMedPubMed CentralGoogle Scholar
 Bloom JD, Silberg JJ, Wilke CO, Drummond DA, Adami C, Arnold FH. Thermodynamic prediction of protein neutrality. Proc Natl Acad Sci USA. 2005; 102:606–11.View ArticlePubMedPubMed CentralGoogle Scholar
Copyright
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.