An Early SARS-CoV-2 Omicron Outbreak in a Dormitory in Saint Petersburg, Russia.

The Omicron variant of SARS-CoV-2 rapidly spread worldwide in late 2021-early 2022, displacing the previously prevalent Delta variant. Before 16 December 2021, community transmission had already been observed in tens of countries globally. However, in Russia, the majority of reported cases at that time had been sporadic and associated with travel. Here, we report an Omicron outbreak at a student dormitory in Saint Petersburg between 16-29 December 2021, which was the earliest known instance of a large-scale community transmission in Russia. Out of the 465 sampled residents of the dormitory, 180 (38.7%) tested PCR-positive. Among the 118 residents for whom the variant had been tested by whole-genome sequencing, 111 (94.1%) were found to carry the Omicron variant. Among these 111 residents, 60 (54.1%) were vaccinated or had reported a previous infection of COVID-19. Phylogenetic analysis confirmed that the outbreak was caused by a single introduction of the BA.1.1 sub-lineage of the Omicron variant. The dormitory-derived clade constituted a significant proportion of BA.1.1 samples in Saint Petersburg and has spread to other regions of Russia and even to other countries. The rapid spread of the Omicron variant in a population with preexisting immunity to previous variants underlines its propensity for immune evasion.

Genomic epidemiology of SARS-CoV-2 in Russia reveals recurring cross-border transmission throughout 2020.

In 2020, SARS-CoV-2 has spread rapidly across the globe, with most nations failing to prevent or substantially delay its introduction. While many countries have imposed some limitations on trans-border passenger traffic, the effect of these measures on the global spread of COVID-19 strains remains unclear. Here, we report an analysis of 3206 whole-genome sequences of SARS-CoV-2 samples from 78 regions of Russia covering the period before the spread of variants of concern (between March and November 2020). We describe recurring imports of multiple COVID-19 strains into Russia throughout this period, giving rise to 457 uniquely Russian transmission lineages, as well as repeated cross-border transmissions of local circulating variants out of Russia. While the phylogenetically inferred rate of cross-border transmissions was somewhat reduced during the period of the most stringent border closure, it still remained high, with multiple inferred imports that each led to detectable spread within the country. These results indicate that partial border closure has had little effect on trans-border transmission of variants, which helps explain the rapid global spread of newly arising SARS-CoV-2 variants throughout the pandemic.

SARS-CoV-2 escape from cytotoxic T cells during long-term COVID-19.

Evolution of SARS-CoV-2 in immunocompromised hosts may result in novel variants with changed properties. While escape from humoral immunity certainly contributes to intra-host evolution, escape from cellular immunity is poorly understood. Here, we report a case of long-term COVID-19 in an immunocompromised patient with non-Hodgkin's lymphoma who received treatment with rituximab and lacked neutralizing antibodies. Over the 318 days of the disease, the SARS-CoV-2 genome gained a total of 40 changes, 34 of which were present by the end of the study period. Among the acquired mutations, 12 reduced or prevented the binding of known immunogenic SARS-CoV-2 HLA class I antigens. By experimentally assessing the effect of a subset of the escape mutations, we show that they resulted in a loss of as much as ~1% of effector CD8 T cell response. Our results indicate that CD8 T cell escape represents a major underappreciated contributor to SARS-CoV-2 evolution in humans.

Human genotype-to-phenotype predictions: Boosting accuracy with nonlinear models.

Genotype-to-phenotype prediction is a central problem of human genetics. In recent years, it has become possible to construct complex predictive models for phenotypes, thanks to the availability of large genome data sets as well as efficient and scalable machine learning tools. In this paper, we make a threefold contribution to this problem. First, we ask if state-of-the-art nonlinear predictive models, such as boosted decision trees, can be more efficient for phenotype prediction than conventional linear models. We find that this is indeed the case if model features include a sufficiently rich set of covariates, but probably not otherwise. Second, we ask if the conventional selection of single nucleotide polymorphisms (SNPs) by genome wide association studies (GWAS) can be replaced by a more efficient procedure, taking into account information in previously selected SNPs. We propose such a procedure, based on a sequential feature importance estimation with decision trees, and show that this approach indeed produced informative SNP sets that are much more compact than when selected with GWAS. Finally, we show that the highest prediction accuracy can ultimately be achieved by ensembling individual linear and nonlinear models. To the best of our knowledge, for some of the phenotypes that we consider (asthma, hypothyroidism), our results are a new state-of-the-art.

Spread of endemic SARS-CoV-2 lineages in Russia before April 2021.

In 2021, the COVID-19 pandemic was characterized by global spread of several lineages with evidence for increased transmissibility. Throughout the pandemic, Russia has remained among the countries with the highest number of confirmed COVID-19 cases, making it a potential hotspot for emergence of novel variants. Here, we show that among the globally significant variants of concern that have spread globally by late 2020, alpha (B.1.1.7), beta (B.1.351) or gamma (P.1), none have been sampled in Russia before the end of 2020. Instead, between summer 2020 and spring 2021, the epidemic in Russia has been characterized by the spread of two lineages that were rare in most other countries: B.1.1.317 and a sublineage of B.1.1 including B.1.1.397 (hereafter, B.1.1.397+). Their frequency has increased concordantly in different parts of Russia. On top of these lineages, in late December 2020, alpha (B.1.1.7) emerged in Russia, reaching a frequency of 17.4% (95% C.I.: 12.0%-24.4%) in March 2021. Additionally, we identify three novel distinct lineages, AT.1, B.1.1.524 and B.1.1.525, that have started to spread, together reaching the frequency of 11.8% (95% C.I.: 7.5%-18.1%) in March 2021. These lineages carry combinations of several notable mutations, including the S:E484K mutation of concern, deletions at a recurrent deletion region of the spike glycoprotein (S:Δ140-142, S:Δ144 or S:Δ136-144), and nsp6:Δ106-108 (also known as ORF1a:Δ3675-3677). Community-based PCR testing indicates that these variants have continued to spread in April 2021, with the frequency of B.1.1.7 reaching 21.7% (95% C.I.: 12.3%-35.6%), and the joint frequency of B.1.1.524 and B.1.1.525, 15.2% (95% C.I.: 7.6%-28.2%). Although these variants have been displaced by the onset of delta variant in May-June 2021, lineages B.1.1.317, B.1.1.397+, AT.1, B.1.1.524 and B.1.1.525 and the combinations of mutations comprising them that are found in other lineages merit monitoring.

The rise and spread of the SARS-CoV-2 AY.122 lineage in Russia.

Delta has outcompeted most preexisting variants of SARS-CoV-2, becoming the globally predominant lineage by mid-2021. Its subsequent evolution has led to the emergence of multiple sublineages, most of which are well-mixed between countries. By contrast, here we show that nearly the entire Delta epidemic in Russia has probably descended from a single import event, or from multiple closely timed imports from a single poorly sampled geographic location. Indeed, over 90 per cent of Delta samples in Russia are characterized by the nsp2:K81N + ORF7a:P45L pair of mutations which is rare outside Russia, putting them in the AY.122 sublineage. The AY.122 lineage was frequent in Russia among Delta samples from the start, and has not increased in frequency in other countries where it has been observed, suggesting that its high prevalence in Russia has probably resulted from a random founder effect rather than a transmission advantage. The apartness of the genetic composition of the Delta epidemic in Russia makes Russia somewhat unusual, although not exceptional, among other countries.

The rise and spread of the SARS-CoV-2 AY.122 lineage in Russia.

BACKGROUND: Delta has outcompeted most preexisting variants of SARS-CoV-2, becoming the globally predominant lineage by mid-2021. Its subsequent evolution has led to emergence of multiple sublineages, many of which are well-mixed between countries. AIM: Here, we aim to study the emergence and spread of the Delta lineage in Russia. METHODS: We use a phylogeographic approach to infer imports of Delta sublineages into Russia, and phylodynamic models to assess the rate of their spread. RESULTS: We show that nearly the entire Delta epidemic in Russia has probably descended from a single import event despite genetic evidence of multiple Delta imports. Indeed, over 90% of Delta samples in Russia are characterized by the nsp2:K81N+ORF7a:P45L pair of mutations which is rare outside Russia, putting them in the AY.122 sublineage. The AY.122 lineage was frequent in Russia among Delta samples from the start, and has not increased in frequency in other countries where it has been observed, suggesting that its high prevalence in Russia has probably resulted from a random founder effect. CONCLUSION: The apartness of the genetic composition of the Delta epidemic in Russia makes Russia somewhat unusual, although not exceptional, among other countries.

SELVa: Simulator of evolution with landscape variation.

Organisms evolve to increase their fitness, a process that may be described as climbing the fitness landscape. However, the fitness landscape of an individual site, i.e., the vector of fitness values corresponding to different variants at this site, can itself change with time due to changes in the environment or substitutions at other epistatically interacting sites. While there exist a number of simulators for modeling different aspects of molecular evolution, very few can accommodate changing landscapes. We present SELVa, the Simulator of Evolution with Landscape Variation, aimed at modeling the substitution process under a changing single-position fitness landscape in a set of evolving lineages that form a phylogeny of arbitrary shape. Written in Java and distributed as an executable jar file, SELVa provides a flexible framework that allows the user to choose from a number of implemented rules governing landscape change.

Accurate fetal variant calling in the presence of maternal cell contamination.

High-throughput sequencing of fetal DNA is a promising and increasingly common method for the discovery of all (or all coding) genetic variants in the fetus, either as part of prenatal screening or diagnosis, or for genetic diagnosis of spontaneous abortions. In many cases, the fetal DNA (from chorionic villi, amniotic fluid, or abortive tissue) can be contaminated with maternal cells, resulting in the mixture of fetal and maternal DNA. This maternal cell contamination (MCC) undermines the assumption, made by traditional variant callers, that each allele in a heterozygous site is covered, on average, by 50% of the reads, and therefore can lead to erroneous genotype calls. We present a panel of methods for reducing the genotyping error in the presence of MCC. All methods start with the output of GATK HaplotypeCaller on the sequencing data for the (contaminated) fetal sample and both of its parents, and additionally rely on information about the MCC fraction (which itself is readily estimated from the high-throughput sequencing data). The first of these methods uses a Bayesian probabilistic model to correct the fetal genotype calls produced by MCC-unaware HaplotypeCaller. The other two methods "learn" the genotype-correction model from examples. We use simulated contaminated fetal data to train and test the models. Using the test sets, we show that all three methods lead to substantially improved accuracy when compared with the original MCC-unaware HaplotypeCaller calls. We then apply the best-performing method to three chorionic villus samples from spontaneously terminated pregnancies.

The miniature genome of a carnivorous plant Genlisea aurea contains a low number of genes and short non-coding sequences.

BACKGROUND: Genlisea aurea (Lentibulariaceae) is a carnivorous plant with unusually small genome size - 63.6 Mb - one of the smallest known among higher plants. Data on the genome sizes and the phylogeny of Genlisea suggest that this is a derived state within the genus. Thus, G. aurea is an excellent model organism for studying evolutionary mechanisms of genome contraction. RESULTS: Here we report sequencing and de novo draft assembly of G. aurea genome. The assembly consists of 10,687 contigs of the total length of 43.4 Mb and includes 17,755 complete and partial protein-coding genes. Its comparison with the genome of Mimulus guttatus, another representative of higher core Lamiales clade, reveals striking differences in gene content and length of non-coding regions. CONCLUSIONS: Genome contraction was a complex process, which involved gene loss and reduction of lengths of introns and intergenic regions, but not intron loss. The gene loss is more frequent for the genes that belong to multigenic families indicating that genetic redundancy is an important prerequisite for genome size reduction.

Organization of physical interactomes as uncovered by network schemas.

Large-scale protein-protein interaction networks provide new opportunities for understanding cellular organization and functioning. We introduce network schemas to elucidate shared mechanisms within interactomes. Network schemas specify descriptions of proteins and the topology of interactions among them. We develop algorithms for systematically uncovering recurring, over-represented schemas in physical interaction networks. We apply our methods to the S. cerevisiae interactome, focusing on schemas consisting of proteins described via sequence motifs and molecular function annotations and interacting with one another in one of four basic network topologies. We identify hundreds of recurring and over-represented network schemas of various complexity, and demonstrate via graph-theoretic representations how more complex schemas are organized in terms of their lower-order constituents. The uncovered schemas span a wide range of cellular activities, with many signaling and transport related higher-order schemas. We establish the functional importance of the schemas by showing that they correspond to functionally cohesive sets of proteins, are enriched in the frequency with which they have instances in the H. sapiens interactome, and are useful for predicting protein function. Our findings suggest that network schemas are a powerful paradigm for organizing, interrogating, and annotating cellular networks.

NetGrep: fast network schema searches in interactomes.

NetGrep (http://genomics.princeton.edu/singhlab/netgrep/) is a system for searching protein interaction networks for matches to user-supplied 'network schemas'. Each schema consists of descriptions of proteins (for example, their molecular functions or putative domains) along with the desired topology and types of interactions among them. Schemas can thus describe domain-domain interactions, signaling and regulatory pathways, or more complex network patterns. NetGrep provides an advanced graphical interface for specifying schemas and fast algorithms for extracting their matches.

Whole-proteome prediction of protein function via graph-theoretic analysis of interaction maps.

MOTIVATION: Determining protein function is one of the most important problems in the post-genomic era. For the typical proteome, there are no functional annotations for one-third or more of its proteins. Recent high-throughput experiments have determined proteome-scale protein physical interaction maps for several organisms. These physical interactions are complemented by an abundance of data about other types of functional relationships between proteins, including genetic interactions, knowledge about co-expression and shared evolutionary history. Taken together, these pairwise linkages can be used to build whole-proteome protein interaction maps. RESULTS: We develop a network-flow based algorithm, FunctionalFlow, that exploits the underlying structure of protein interaction maps in order to predict protein function. In cross-validation testing on the yeast proteome, we show that FunctionalFlow has improved performance over previous methods in predicting the function of proteins with few (or no) annotated protein neighbors. By comparing several methods that use protein interaction maps to predict protein function, we demonstrate that FunctionalFlow performs well because it takes advantage of both network topology and some measure of locality. Finally, we show that performance can be improved substantially as we consider multiple data sources and use them to create weighted interaction networks. AVAILABILITY: http://compbio.cs.princeton.edu/function