The protein domains of vertebrate species in which selection is more effective have greater intrinsic structural disorder.

The nearly neutral theory of molecular evolution posits variation among species in the effectiveness of selection. In an idealized model, the census population size determines both this minimum magnitude of the selection coefficient required for deleterious variants to be reliably purged, and the amount of neutral diversity. Empirically, an "effective population size" is often estimated from the amount of putatively neutral genetic diversity and is assumed to also capture a species' effectiveness of selection. A potentially more direct measure of the effectiveness of selection is the degree to which selection maintains preferred codons. However, past metrics that compare codon bias across species are confounded by among-species variation in %GC content and/or amino acid composition. Here we propose a new Codon Adaptation Index of Species (CAIS), based on Kullback-Leibler divergence, that corrects for both confounders. We demonstrate the use of CAIS correlations, as well as the Effective Number of Codons, to show that the protein domains of more highly adapted vertebrate species evolve higher intrinsic structural disorder.

Order of amino acid recruitment into the genetic code resolved by Last Universal Common Ancestor's protein domains.

The current "consensus" order in which amino acids were added to the genetic code is based on potentially biased criteria such as absence of sulfur-containing amino acids from the Urey-Miller experiment which lacked sulfur. Even if inferred perfectly, abiotic abundance might not reflect abundance in the organisms in which the genetic code evolved. Here, we instead exploit the fact that proteins that emerged prior to the genetic code's completion are likely enriched in early amino acids and depleted in late amino acids. We identify the most ancient protein-coding sequences born prior to the archaeal-bacterial split. Amino acid usage in protein sequences whose ancestors date back to a single homolog in the Last Universal Common Ancestor (LUCA) largely matches the consensus order. However, our findings indicate that metal-binding (cysteine and histidine) and sulfur-containing (cysteine and methionine) amino acids were added to the genetic code much earlier than previously thought. Surprisingly, even more ancient protein sequences - those that had already diversified into multiple distinct copies in LUCA - show a different pattern to single copy LUCA sequences: significantly less depleted in the late amino acids tryptophan and tyrosine, and enriched rather than depleted in phenylalanine. This is compatible with at least some of these sequences predating the current genetic code. Their distinct enrichment patterns thus provide hints about earlier, alternative genetic codes.

Background Selection From Unlinked Sites Causes Nonindependent Evolution of Deleterious Mutations.

Background selection describes the reduction in neutral diversity caused by selection against deleterious alleles at other loci. It is typically assumed that the purging of deleterious alleles affects linked neutral variants, and indeed simulations typically only treat a genomic window. However, background selection at unlinked loci also depresses neutral diversity. In agreement with previous analytical approximations, in our simulations of a human-like genome with a realistically high genome-wide deleterious mutation rate, the effects of unlinked background selection exceed those of linked background selection. Background selection reduces neutral genetic diversity by a factor that is independent of census population size. Outside of genic regions, the strength of background selection increases with the mean selection coefficient, contradicting the linked theory but in agreement with the unlinked theory. Neutral diversity within genic regions is fairly independent of the strength of selection. Deleterious genetic load among haploid individuals is underdispersed, indicating nonindependent evolution of deleterious mutations. Empirical evidence for underdispersion was previously interpreted as evidence for global epistasis, but we recover it from a non-epistatic model.

Combatting SARS-CoV-2 With Digital Contact Tracing and Notification: Navigating Six Points of Failure.

Digital contact tracing and notification were initially hailed as promising strategies to combat SARS-CoV-2; however, in most jurisdictions, they did not live up to their promise. To avert a given transmission event, both parties must have adopted the technology, it must detect the contact, the primary case must be promptly diagnosed, notifications must be triggered, and the secondary case must change their behavior to avoid the focal tertiary transmission event. If we approximate these as independent events, achieving a 26% reduction in the effective reproduction number Rt would require an 80% success rate at each of these 6 points of failure. Here, we review the 6 failure rates experienced by a variety of digital contact tracing and contact notification schemes, including Singapore's TraceTogether, India's Aarogya Setu, and leading implementations of the Google Apple Exposure Notification system. This leads to a number of recommendations, for example, that the narrative be framed in terms of user autonomy rather than user privacy, and that tracing/notification apps be multifunctional and integrated with testing, manual contact tracing, and the gathering of critical scientific data.

Human deleterious mutation rate implies high fitness variance, with declining mean fitness compensated by rarer beneficial mutations of larger effect.

Each new human has an expected Ud = 2 - 10 new deleterious mutations. This deluge of deleterious mutations cannot all be purged, and therefore accumulate in a declining fitness ratchet. Using a novel simulation framework designed to efficiently handle genome-wide linkage disequilibria across many segregating sites, we find that rarer, beneficial mutations of larger effect are sufficient to compensate fitness declines due to the fixation of many slightly deleterious mutations. Drift barrier theory posits a similar asymmetric pattern of fixations to explain ratcheting genome size and complexity, but in our theory, the cause is Ud > 1 rather than small population size. In our simulations, Ud ~2 - 10 generates high within-population variance in relative fitness; two individuals will typically differ in fitness by 15-40%. Ud ~2 - 10 also slows net adaptation by ~13%-39%. Surprisingly, fixation rates are more sensitive to changes in the beneficial than the deleterious mutation rate, e.g. a 10% increase in overall mutation rate leads to faster adaptation; this puts to rest dysgenic fears about increasing mutation rates due to rising paternal age.

Differential Retention of Pfam Domains Contributes to Long-term Evolutionary Trends.

Protein domains that emerged more recently in evolution have a higher structural disorder and greater clustering of hydrophobic residues along the primary sequence. It is hard to explain how selection acting via descent with modification could act so slowly as not to saturate over the extraordinarily long timescales over which these trends persist. Here, we hypothesize that the trends were created by a higher level of selection that differentially affects the retention probabilities of protein domains with different properties. This hypothesis predicts that loss rates should depend on disorder and clustering trait values. To test this, we inferred loss rates via maximum likelihood for animal Pfam domains, after first performing a set of stringent quality control methods to reduce annotation errors. Intermediate trait values, matching those of ancient domains, are associated with the lowest loss rates, making our results difficult to explain with reference to previously described homology detection biases. Simulations confirm that effect sizes are of the right magnitude to produce the observed long-term trends. Our results support the hypothesis that differential domain loss slowly weeds out those protein domains that have nonoptimal levels of disorder and clustering. The same preferences also shape the differential diversification of Pfam domains, thereby further impacting proteome composition.

Random peptides rich in small and disorder-promoting amino acids are less likely to be harmful.

Proteins are the workhorses of the cell, yet they carry great potential for harm via misfolding and aggregation. Despite the dangers, proteins are sometimes born de novo from non-coding DNA. Proteins are more likely to be born from non-coding regions that produce peptides that do little to no harm when translated than from regions that produce harmful peptides. To investigate which newborn proteins are most likely to "first, do no harm", we estimate fitnesses from an experiment that competed Escherichia coli lineages that each expressed a unique random peptide. A variety of peptide metrics significantly predict lineage fitness, but this predictive power stems from simple amino acid frequencies rather than the ordering of amino acids. Amino acids that are smaller and that promote intrinsic structural disorder have more benign fitness effects. We validate that the amino acids that indicate benign effects in random peptides expressed in E. coli also do so in an independent dataset of random N-terminal tags in which it is possible to control for expression level. The same amino acids are also enriched in young animal proteins.

nQMaker: Estimating Time Nonreversible Amino Acid Substitution Models.

Amino acid substitution models are a key component in phylogenetic analyses of protein sequences. All commonly used amino acid models available to date are time-reversible, an assumption designed for computational convenience but not for biological reality. Another significant downside to time-reversible models is that they do not allow inference of rooted trees without outgroups. In this article, we introduce a maximum likelihood approach nQMaker, an extension of the recently published QMaker method, that allows the estimation of time nonreversible amino acid substitution models and rooted phylogenetic trees from a set of protein sequence alignments. We show that the nonreversible models estimated with nQMaker are a much better fit to empirical alignments than pre-existing reversible models, across a wide range of data sets including mammals, birds, plants, fungi, and other taxa, and that the improvements in model fit scale with the size of the data set. Notably, for the recently published plant and bird trees, these nonreversible models correctly recovered the commonly estimated root placements with very high-statistical support without the need to use an outgroup. We provide nQMaker as an easy-to-use feature in the IQ-TREE software (http://www.iqtree.org), allowing users to estimate nonreversible models and rooted phylogenies from their own protein data sets. The data sets and scripts used in this article are available at https://doi.org/10.5061/dryad.3tx95x6hx. [amino acid sequence analyses; amino acid substitution models; maximum likelihood model estimation; nonreversible models; phylogenetic inference; reversible models.].

Quantifying SARS-CoV-2 Infection Risk Within the Google/Apple Exposure Notification Framework to Inform Quarantine Recommendations.

Most early Bluetooth-based exposure notification apps use three binary classifications to recommend quarantine following SARS-CoV-2 exposure: a window of infectiousness in the transmitter, ≥15 minutes duration, and Bluetooth attenuation below a threshold. However, Bluetooth attenuation is not a reliable measure of distance, and infection risk is not a binary function of distance, nor duration, nor timing. We model uncertainty in the shape and orientation of an exhaled virus-containing plume and in inhalation parameters, and measure uncertainty in distance as a function of Bluetooth attenuation. We calculate expected dose by combining this with estimated infectiousness based on timing relative to symptom onset. We calibrate an exponential dose-response curve based on infection probabilities of household contacts. The probability of current or future infectiousness, conditioned on how long postexposure an exposed individual has been symptom-free, decreases during quarantine, with shape determined by incubation periods, proportion of asymptomatic cases, and asymptomatic shedding durations. It can be adjusted for negative test results using Bayes' theorem. We capture a 10-fold range of risk using six infectiousness values, 11-fold range using three Bluetooth attenuation bins, ∼sixfold range from exposure duration given the 30 minute duration cap imposed by the Google/Apple v1.1, and ∼11-fold between the beginning and end of 14 day quarantine. Public health authorities can either set a threshold on initial infection risk to determine 14-day quarantine onset, or on the conditional probability of current and future infectiousness conditions to determine both quarantine and duration.

Differences in evolutionary accessibility determine which equally effective regulatory motif evolves to generate pulses.

Transcriptional regulatory networks (TRNs) are enriched for certain "motifs." Motif usage is commonly interpreted in adaptationist terms, i.e., that the optimal motif evolves. But certain motifs can also evolve more easily than others. Here, we computationally evolved TRNs to produce a pulse of an effector protein. Two well-known motifs, type 1 incoherent feed-forward loops (I1FFLs) and negative feedback loops (NFBLs), evolved as the primary solutions. The relative rates at which these two motifs evolve depend on selection conditions, but under all conditions, either motif achieves similar performance. I1FFLs generally evolve more often than NFBLs. Selection for a tall pulse favors NFBLs, while selection for a fast response favors I1FFLs. I1FFLs are more evolutionarily accessible early on, before the effector protein evolves high expression; when NFBLs subsequently evolve, they tend to do so from a conjugated I1FFL-NFBL genotype. In the empirical S. cerevisiae TRN, output genes of NFBLs had higher expression levels than those of I1FFLs. These results suggest that evolutionary accessibility, and not relative functionality, shapes which motifs evolve in TRNs, and does so as a function of the expression levels of particular genes.

The economic value of quarantine is higher at lower case prevalence, with quarantine justified at lower risk of infection.

Some infectious diseases, such as COVID-19 or the influenza pandemic of 1918, are so harmful that they justify broad-scale social distancing. Targeted quarantine can reduce the amount of indiscriminate social distancing needed to control transmission. Finding the optimal balance between targeted versus broad-scale policies can be operationalized by minimizing the total amount of social isolation needed to achieve a target reproductive number. Optimality is achieved by quarantining on the basis of a risk threshold that depends strongly on current disease prevalence, suggesting that very different disease control policies should be used at different times or places. Aggressive quarantine is warranted given low disease prevalence, while populations with a higher base rate of infection should rely more on social distancing by all. The total value of a quarantine policy rises as case counts fall, is relatively insensitive to vaccination unless the vaccinated are exempt from distancing policies, and is substantially increased by the availability of modestly more information about individual risk of infectiousness.

Universal and taxon-specific trends in protein sequences as a function of age.

Extant protein-coding sequences span a huge range of ages, from those that emerged only recently to those present in the last universal common ancestor. Because evolution has had less time to act on young sequences, there might be 'phylostratigraphy' trends in any properties that evolve slowly with age. A long-term reduction in hydrophobicity and hydrophobic clustering was found in previous, taxonomically restricted studies. Here we perform integrated phylostratigraphy across 435 fully sequenced species, using sensitive HMM methods to detect protein domain homology. We find that the reduction in hydrophobic clustering is universal across lineages. However, only young animal domains have a tendency to have higher structural disorder. Among ancient domains, trends in amino acid composition reflect the order of recruitment into the genetic code, suggesting that the composition of the contemporary descendants of ancient sequences reflects amino acid availability during the earliest stages of life, when these sequences first emerged.

Mutation bias can shape adaptation in large asexual populations experiencing clonal interference.

The extended evolutionary synthesis invokes a role for development in shaping adaptive evolution, which in population genetics terms corresponds to mutation-biased adaptation. Critics have claimed that clonal interference makes mutation-biased adaptation rare. We consider the behaviour of two simultaneously adapting traits, one with larger mutation rate U, the other with larger selection coefficient s, using asexual travelling wave models. We find that adaptation is dominated by whichever trait has the faster rate of adaptation v in isolation, with the other trait subject to evolutionary stalling. Reviewing empirical claims for mutation-biased adaptation, we find that not all occur in the 'origin-fixation' regime of population genetics where v is only twice as sensitive to s as to U. In some cases, differences in U are at least ten to twelve times larger than differences in s, as needed to cause mutation-biased adaptation even in the 'multiple mutations' regime. Surprisingly, when U > s in the 'diffusive-mutation' regime, the required sensitivity ratio is also only two, despite pervasive clonal interference. Given two traits with identical v, the benefit of having higher s is surprisingly small, occurring largely when one trait is at the boundary between the origin-fixation and multiple mutations regimes.

Only a Single Taxonomically Restricted Gene Family in the Drosophila melanogaster Subgroup Can Be Identified with High Confidence.

Taxonomically restricted genes (TRGs) are genes that are present only in one clade. Protein-coding TRGs may evolve de novo from previously noncoding sequences: functional ncRNA, introns, or alternative reading frames of older protein-coding genes, or intergenic sequences. A major challenge in studying de novo genes is the need to avoid both false-positives (nonfunctional open reading frames and/or functional genes that did not arise de novo) and false-negatives. Here, we search conservatively for high-confidence TRGs as the most promising candidates for experimental studies, ensuring functionality through conservation across at least two species, and ensuring de novo status through examination of homologous noncoding sequences. Our pipeline also avoids ascertainment biases associated with preconceptions of how de novo genes are born. We identify one TRG family that evolved de novo in the Drosophila melanogaster subgroup. This TRG family contains single-copy genes in Drosophila simulans and Drosophila sechellia. It originated in an intron of a well-established gene, sharing that intron with another well-established gene upstream. These TRGs contain an intron that predates their open reading frame. These genes have not been previously reported as de novo originated, and to our knowledge, they are the best Drosophila candidates identified so far for experimental studies aimed at elucidating the properties of de novo genes.

Readthrough Errors Purge Deleterious Cryptic Sequences, Facilitating the Birth of Coding Sequences.

De novo protein-coding innovations sometimes emerge from ancestrally noncoding DNA, despite the expectation that translating random sequences is overwhelmingly likely to be deleterious. The "preadapting selection" hypothesis claims that emergence is facilitated by prior, low-level translation of noncoding sequences via molecular errors. It predicts that selection on polypeptides translated only in error is strong enough to matter and is strongest when erroneous expression is high. To test this hypothesis, we examined noncoding sequences located downstream of stop codons (i.e., those potentially translated by readthrough errors) in Saccharomyces cerevisiae genes. We identified a class of "fragile" proteins under strong selection to reduce readthrough, which are unlikely substrates for co-option. Among the remainder, sequences showing evidence of readthrough translation, as assessed by ribosome profiling, encoded C-terminal extensions with higher intrinsic structural disorder, supporting the preadapting selection hypothesis. The cryptic sequences beyond the stop codon, rather than spillover effects from the regular C-termini, are primarily responsible for the higher disorder. Results are robust to controlling for the fact that stronger selection also reduces the length of C-terminal extensions. These findings indicate that selection acts on 3' UTRs in Saccharomyces cerevisiae to purge potentially deleterious variants of cryptic polypeptides, acting more strongly in genes that experience more readthrough errors.

Evolution Rapidly Optimizes Stability and Aggregation in Lattice Proteins Despite Pervasive Landscape Valleys and Mazes.

The "fitness" landscapes of genetic sequences are characterized by high dimensionality and "ruggedness" due to sign epistasis. Ascending from low to high fitness on such landscapes can be difficult because adaptive trajectories get stuck at low-fitness local peaks. Compounding matters, recent theoretical arguments have proposed that extremely long, winding adaptive paths may be required to reach even local peaks: a "maze-like" landscape topography. The extent to which peaks and mazes shape the mode and tempo of evolution is poorly understood, due to empirical limitations and the abstractness of many landscape models. We explore the prevalence, scale, and evolutionary consequences of landscape mazes in a biophysically grounded computational model of protein evolution that captures the "frustration" between "stability" and aggregation propensity. Our stability-aggregation landscape exhibits extensive sign epistasis and local peaks galore. Although this frequently obstructs adaptive ascent to high fitness and virtually eliminates reproducibility of evolutionary outcomes, many adaptive paths do successfully complete the ascent from low to high fitness, with hydrophobicity a critical mediator of success. These successful paths exhibit maze-like properties on a global landscape scale, in which taking an indirect path helps to avoid low-fitness local peaks. This delicate balance of "hard but possible" adaptation could occur more broadly in other biological settings where competing interactions and frustration are important.

Biomarkers for Aging Identified in Cross-sectional Studies Tend to Be Non-causative.

Biomarkers are important tools for diagnosis, prognosis, and identification of the causal factors of physiological conditions. Biomarkers are typically identified by correlating biological measurements with the status of a condition in a sample of subjects. Cross-sectional studies sample subjects at a single timepoint, whereas longitudinal studies follow a cohort through time. Identifying biomarkers of aging is subject to unique challenges. Individuals who age faster have intrinsically higher mortality rates and so are preferentially lost over time, in a phenomenon known as cohort selection. In this article, we use simulations to show that cohort selection biases cross-sectional analysis away from identifying causal loci of aging, to the point where cross-sectional studies are less likely to identify loci that cause aging than if loci had been chosen at random. We go on to show this bias can be corrected by incorporating correlates of mortality identified from longitudinal studies, allowing cross-sectional studies to effectively identify the causal factors of aging.

High Transcriptional Error Rates Vary as a Function of Gene Expression Level.

Errors in gene transcription can be costly, and organisms have evolved to prevent their occurrence or mitigate their costs. The simplest interpretation of the drift barrier hypothesis suggests that species with larger population sizes would have lower transcriptional error rates. However, Escherichia coli seems to have a higher transcriptional error rate than species with lower effective population sizes, for example Saccharomyces cerevisiae. This could be explained if selection in E. coli were strong enough to maintain adaptations that mitigate the consequences of transcriptional errors through robustness, on a gene by gene basis, obviating the need for low transcriptional error rates and associated costs of global proofreading. Here, we note that if selection is powerful enough to evolve local robustness, selection should also be powerful enough to locally reduce error rates. We therefore predict that transcriptional error rates will be lower in highly abundant proteins on which selection is strongest. However, we only expect this result when error rates are high enough to significantly impact fitness. As expected, we find such a relationship between expression and transcriptional error rate for non-C→U errors in E. coli (especially G→A), but not in S. cerevisiae. We do not find this pattern for C→U changes in E. coli, presumably because most deamination events occurred during sample preparation, but do for C→U changes in S. cerevisiae, supporting the interpretation that C→U error rates estimated with an improved protocol, and which occur at rates comparable with E. coli non-C→U errors, are biological.

Feed-forward regulation adaptively evolves via dynamics rather than topology when there is intrinsic noise.

In transcriptional regulatory networks (TRNs), a canonical 3-node feed-forward loop (FFL) is hypothesized to evolve to filter out short spurious signals. We test this adaptive hypothesis against a novel null evolutionary model. Our mutational model captures the intrinsically high prevalence of weak affinity transcription factor binding sites. We also capture stochasticity and delays in gene expression that distort external signals and intrinsically generate noise. Functional FFLs evolve readily under selection for the hypothesized function but not in negative controls. Interestingly, a 4-node "diamond" motif also emerges as a short spurious signal filter. The diamond uses expression dynamics rather than path length to provide fast and slow pathways. When there is no idealized external spurious signal to filter out, but only internally generated noise, only the diamond and not the FFL evolves. While our results support the adaptive hypothesis, we also show that non-adaptive factors, including the intrinsic expression dynamics, matter.