A new profiling approach for DNA sequences based on the nucleotides' physicochemical features for accurate analysis of SARS-CoV-2 genomes.

BACKGROUND: The prevalence of the COVID-19 disease in recent years and its widespread impact on mortality, as well as various aspects of life around the world, has made it important to study this disease and its viral cause. However, very long sequences of this virus increase the processing time, complexity of calculation, and memory consumption required by the available tools to compare and analyze the sequences. RESULTS: We present a new encoding method, named PC-mer, based on the k-mer and physic-chemical properties of nucleotides. This method minimizes the size of encoded data by around 2 k times compared to the classical k-mer based profiling method. Moreover, using PC-mer, we designed two tools: 1) a machine-learning-based classification tool for coronavirus family members with the ability to recive input sequences from the NCBI database, and 2) an alignment-free computational comparison tool for calculating dissimilarity scores between coronaviruses at the genus and species levels. CONCLUSIONS: PC-mer achieves 100% accuracy despite the use of very simple classification algorithms based on Machine Learning. Assuming dynamic programming-based pairwise alignment as the ground truth approach, we achieved a degree of convergence of more than 98% for coronavirus genus-level sequences and 93% for SARS-CoV-2 sequences using PC-mer in the alignment-free classification method. This outperformance of PC-mer suggests that it can serve as a replacement for alignment-based approaches in certain sequence analysis applications that rely on similarity/dissimilarity scores, such as searching sequences, comparing sequences, and certain types of phylogenetic analysis methods that are based on sequence comparison.

Evolution of coronavirus frameshifting elements: Competing stem networks explain conservation and variability.

The frameshifting RNA element (FSE) in coronaviruses (CoVs) regulates the programmed -1 ribosomal frameshift (-1 PRF) mechanism common to many viruses. The FSE is of particular interest as a promising drug candidate. Its associated pseudoknot or stem loop structure is thought to play a large role in frameshifting and thus viral protein production. To investigate the FSE structural evolution, we use our graph theory-based methods for representing RNA secondary structures in the RNA-As-Graphs (RAG) framework to calculate conformational landscapes of viral FSEs with increasing sequence lengths for representative 10 Alpha and 13 Beta-CoVs. By following length-dependent conformational changes, we show that FSE sequences encode many possible competing stems which in turn favor certain FSE topologies, including a variety of pseudoknots, stem loops, and junctions. We explain alternative competing stems and topological FSE changes by recurring patterns of mutations. At the same time, FSE topology robustness can be understood by shifted stems within different sequence contexts and base pair coevolution. We further propose that the topology changes reflected by length-dependent conformations contribute to tuning the frameshifting efficiency. Our work provides tools to analyze virus sequence/structure correlations, explains how sequence and FSE structure have evolved for CoVs, and provides insights into potential mutations for therapeutic applications against a broad spectrum of CoV FSEs by targeting key sequence/structural transitions.

Intragenomic rearrangements involving 5'-untranslated region segments in SARS-CoV-2, other betacoronaviruses, and alphacoronaviruses.

BACKGROUND: Variation of the betacoronavirus SARS-CoV-2 has been the bane of COVID-19 control. Documented variation includes point mutations, deletions, insertions, and recombination among closely or distantly related coronaviruses. Here, we describe yet another aspect of genome variation by beta- and alphacoronaviruses that was first documented in an infectious isolate of the betacoronavirus SARS-CoV-2, obtained from 3 patients in Hong Kong that had a 5'-untranslated region segment at the end of the ORF6 gene that in its new location translated into an ORF6 protein with a predicted modified carboxyl terminus. While comparing the amino acid sequences of translated ORF8 genes in the GenBank database, we found a subsegment of the same 5'-UTR-derived amino acid sequence modifying the distal end of ORF8 of an isolate from the United States and decided to carry out a systematic search. METHODS: Using the nucleotide and in the case of SARS-CoV-2 also the translated amino acid sequence in three reading frames of the genomic termini of coronaviruses as query sequences, we searched for 5'-UTR sequences in regions other than the 5'-UTR in SARS-CoV-2 and reference strains of alpha-, beta-, gamma-, and delta-coronaviruses. RESULTS: We here report numerous genomic insertions of 5'-untranslated region sequences into coding regions of SARS-CoV-2, other betacoronaviruses, and alphacoronaviruses, but not delta- or gammacoronaviruses. To our knowledge this is the first systematic description of such insertions. In many cases, these insertions would change viral protein sequences and further foster genomic flexibility and viral adaptability through insertion of transcription regulatory sequences in novel positions within the genome. Among human Embecorivus betacoronaviruses, for instance, from 65% to all of the surveyed sequences in publicly available databases contain inserted 5'-UTR sequences. CONCLUSION: The intragenomic rearrangements involving 5'-untranslated region sequences described here, which in several cases affect highly conserved genes with a low propensity for recombination, may underlie the generation of variants homotypic with those of concern or interest and with potentially differing pathogenic profiles. Intragenomic rearrangements thus add to our appreciation of how variants of SARS-CoV-2 and other beta- and alphacoronaviruses may arise.

Integrated mRNA sequence optimization using deep learning.

The coronavirus disease of 2019 pandemic has catalyzed the rapid development of mRNA vaccines, whereas, how to optimize the mRNA sequence of exogenous gene such as severe acute respiratory syndrome coronavirus 2 spike to fit human cells remains a critical challenge. A new algorithm, iDRO (integrated deep-learning-based mRNA optimization), is developed to optimize multiple components of mRNA sequences based on given amino acid sequences of target protein. Considering the biological constraints, we divided iDRO into two steps: open reading frame (ORF) optimization and 5' untranslated region (UTR) and 3'UTR generation. In ORF optimization, BiLSTM-CRF (bidirectional long-short-term memory with conditional random field) is employed to determine the codon for each amino acid. In UTR generation, RNA-Bart (bidirectional auto-regressive transformer) is proposed to output the corresponding UTR. The results show that the optimized sequences of exogenous genes acquired the pattern of human endogenous gene sequence. In experimental validation, the mRNA sequence optimized by our method, compared with conventional method, shows higher protein expression. To the best of our knowledge, this is the first study by introducing deep-learning methods to integrated mRNA sequence optimization, and these results may contribute to the development of mRNA therapeutics.

Unsupervised outlier detection applied to SARS-CoV-2 nucleotide sequences can identify sequences of common variants and other variants of interest.

As of June 2022, the GISAID database contains more than 11 million SARS-CoV-2 genomes, including several thousand nucleotide sequences for the most common variants such as delta or omicron. These SARS-CoV-2 strains have been collected from patients around the world since the beginning of the pandemic. We start by assessing the similarity of all pairs of nucleotide sequences using the Jaccard index and principal component analysis. As shown previously in the literature, an unsupervised cluster analysis applied to the SARS-CoV-2 genomes results in clusters of sequences according to certain characteristics such as their strain or their clade. Importantly, we observe that nucleotide sequences of common variants are often outliers in clusters of sequences stemming from variants identified earlier on during the pandemic. Motivated by this finding, we are interested in applying outlier detection to nucleotide sequences. We demonstrate that nucleotide sequences of common variants (such as alpha, delta, or omicron) can be identified solely based on a statistical outlier criterion. We argue that outlier detection might be a useful surveillance tool to identify emerging variants in real time as the pandemic progresses.

Description of the novel HLA-C allele, HLA-C*12:376, identified in a deceased COVID-19 patient.

The new allele HLA-C*12:376 showed one nonsynonymous nucleotide difference compared with the C*12:03:01:01 allele in codon 30.

Capturing SARS-CoV-2 from patient samples with low viral abundance: a comparative analysis.

Since the beginning of the SARS-CoV-2 coronavirus pandemic, genome sequencing is essential to monitor viral mutations over time and by territory. This need for complete genetic information is further reinforced by the rapid spread of variants of concern. In this paper, we assess the ability of the hybridization technique, Capture-Seq, to detect the SARS-CoV-2 genome, either partially or in its integrity on patients samples. We studied 20 patient nasal swab samples broken down into five series of four samples of equivalent viral load from CT25 to CT36+ . For this, we tested 3 multi-virus panel as well as 2 SARS-CoV-2 only panels. The panels were chosen based on their specificity, global or specific, as well as their technological difference in the composition of the probes: ssRNA, ssDNA and dsDNA. The multi-virus panels are able to capture high-abundance targets but fail to capture the lowest-abundance targets, with a high percentage of off-target reads corresponding to the abundance of the host sequences. Both SARS-CoV-2-only panels were very effective, with high percentage of reads corresponding to the target. Overall, capture followed by sequencing is very effective for the study of SARS-CoV-2 in low-abundance patient samples and is suitable for samples with CT values up to 35.

Tracking the progressive spread of the SARS-CoV-2 Omicron variant in Italy, December 2021 to January 2022.

BackgroundThe SARS-CoV-2 variant of concern Omicron was first detected in Italy in November 2021.AimTo comprehensively describe Omicron spread in Italy in the 2 subsequent months and its impact on the overall SARS-CoV-2 circulation at population level.MethodsWe analyse data from four genomic surveys conducted across the country between December 2021 and January 2022. Combining genomic sequencing results with epidemiological records collated by the National Integrated Surveillance System, the Omicron reproductive number and exponential growth rate are estimated, as well as SARS-CoV-2 transmissibility.ResultsOmicron became dominant in Italy less than 1 month after its first detection, representing on 3 January 76.9-80.2% of notified SARS-CoV-2 infections, with a doubling time of 2.7-3.3 days. As of 17 January 2022, Delta variant represented < 6% of cases. During the Omicron expansion in December 2021, the estimated mean net reproduction numbers respectively rose from 1.15 to a maximum of 1.83 for symptomatic cases and from 1.14 to 1.36 for hospitalised cases, while remaining relatively stable, between 0.93 and 1.21, for cases needing intensive care. Despite a reduction in relative proportion, Delta infections increased in absolute terms throughout December contributing to an increase in hospitalisations. A significant reproduction numbers' decline was found after mid-January, with average estimates dropping below 1 between 10 and 16 January 2022.ConclusionEstimates suggest a marked growth advantage of Omicron compared with Delta variant, but lower disease severity at population level possibly due to residual immunity against severe outcomes acquired from vaccination and prior infection.

Statistical analysis supports UTR (untranslated region) deletion theory in SARS-CoV-2.

It was noticed that the mortality rate of SARS-CoV-2 infection experienced a significant declination in the early stage of the epidemic. We suspect that the sharp deterioration of virus toxicity is related to the deletion of the untranslated region (UTR) of the virus genome. It was found that the genome length of SARS-CoV-2 engaged a significant truncation due to UTR deletion after a mega-sequence analysis. Sequence similarity analysis further indicated that short UTR strains originated from its long UTR ancestors after an irreversible deletion. A good correlation was discovered between genome length and mortality, which demonstrated that the deletion of the virus UTR significantly affected the toxicity of the virus. This correlation was further confirmed in a significance analysis of the genetic influence on the clinical outcomes. The viral genome length of hospitalized patients was significantly more extensive than that of asymptomatic patients. In contrast, the viral genome length of asymptomatic was considerably longer than that of ordinary patients with symptoms. A genome-level mutation scanning was performed to systematically evaluate the influence of mutations at each position on virulence. The results indicated that UTR deletion was the primary driving force in alternating virus virulence in the early evolution. In the end, we proposed a mathematical model to explain why this UTR deletion was not continuous.

Validation of reduced S-gene target performance and failure for rapid surveillance of SARS-CoV-2 variants.

SARS-CoV-2, the virus that causes COVID-19, has many variants capable of rapid transmission causing serious illness. Timely surveillance of new variants is essential for an effective public health response. Ensuring availability and access to diagnostic and molecular testing is key to this type of surveillance. This study utilized reverse transcription polymerase chain reaction (RT-PCR) and whole genome sequencing results from COVID-19-positive patient samples obtained through a collaboration between Aegis Sciences Corporation and Walgreens Pharmacy that has conducted more than 8.5 million COVID-19 tests at ~5,200 locations across the United States and Puerto Rico. Viral evolution of SARS-CoV-2 can lead to mutations in the S-gene that cause reduced or failed S-gene amplification in diagnostic PCR tests. These anomalies, labeled reduced S-gene target performance (rSGTP) and S-gene target failure (SGTF), are characteristic of variants carrying the del69-70 mutation, such as Alpha and Omicron (B.1.1.529, BA.1, and BA.1.1) lineages. This observation has been validated by whole genome sequencing and can provide presumptive lineage data following completion of diagnostic PCR testing in 24-48 hours from collection. Active surveillance of trends in PCR and sequencing results is key to the identification of changes in viral transmission and emerging variants. This study shows that rSGTP and SGTF can be utilized for near real-time tracking and surveillance of SARS-CoV-2 variants, and is superior to the use of SGTF alone due to the significant proportion of Alpha and Omicron (B.1.1.529, BA.1, and BA.1.1) lineages known to carry the del69-70 mutation and observed to have S-gene amplification. Adopting new tools and techniques to both diagnose acute infections and expedite identification of emerging variants is critical to supporting public health.

Discovery of a tRNA-like base sequence in the coronavirus genome and possible mechanism of action.

BACKGROUND: The question of whether the coronavirus genome contain as-yetununderstood genetic component. PURPOSE (OBJECTIVE): Elucidate the novel functions of the discovered tRNA-like base sequence and lead to the development of novel therapeutic agents. METHODS: A novel tRNA-like base sequence was found in the sequences complementary to the genomes of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and SARS-CoV. By comparing mutations in the tRNA-like base sequences of these two viruses, it was found that base pairing in the cloverleaf model of SARS-CoV-2 was more robust than that of SARS-CoV. RESULTS: The results of homology search between a short sequence of the coronavirus tRNA-like base sequence and human genes suggest that the molecule produced by this novel tRNA-like sequence may be involved in the splicing of human messenger RNA. CONCLUSIONS: Experimental molecular evidence of the tRNA-like base sequence discovered in this study is urgently needed.

Development of multiplex S-gene-targeted RT-PCR for rapid identification of SARS-CoV-2 variants by extended S-gene target failure.

BACKGROUND: Tracking SARS-CoV-2 variants of concern (VOC) by genomic sequencing is time-consuming. The rapid screening of VOCs is necessary for clinical laboratories. In this study, we developed a rapid screening method based on multiplex RT-PCR by extended S-gene target failure (eSGTF), a false negative result caused by S-gene mutations. METHODS: Three S-gene target (SGT) regions (SGT1, codons 65-72; SGT2, codons 152-159; and SGT3, codons 370-377) and an N-gene region (for internal control) were detected in single-tube. Four types of VOC (Alpha, Delta, Omicron BA.1, and Omicron BA.2) are classified by positive/negative patterns of 3 S-gene regions (eSGTF pattern). RESULTS: The eSGTF patterns of VOCs were as follows (SGT1, SGT2, SGT3; P, positive; N, negative): Alpha, NPP; Delta, PNP; Omicron BA.1, NPN pattern; and Omicron BA.2, PPN. As compared with the S-gene sequencing, eSGTF patterns were identical to the specific VOCs (concordance rate = 96.7%, N = 206/213). Seven samples with discordant results had a minor mutation in the probe binding region. The epidemics of VOCs estimated by eSGTF patterns were similar to those in Japan. CONCLUSIONS: Multiplex RT-PCR and eSGTF patterns enable high-throughput screening of VOCs. It will be useful for the rapid determination of VOCs in clinical laboratories.

kmer2vec: A Novel Method for Comparing DNA Sequences by word2vec Embedding.

The comparison of DNA sequences is of great significance in genomics analysis. Although the traditional multiple sequence alignment (MSA) method is popularly used for evolutionary analysis, optimally aligning k sequences becomes computationally intractable when k increases due to the intrinsic computational complexity of MSA. Despite numerous k-mer alignment-free methods being proposed, the existing k-mer alignment-free methods may not truly capture the contextual structures of the sequences. In this study, we present a novel k-mer contextual alignment-free method (called kmer2vec), in which the sequence k-mers are semantically embedded to word2vec vectors, an essential technique in natural language processing. Consequently, the method converts each DNA/RNA sequence into a point in the word2vec high-dimensional space and compares DNA sequences in the space. Because the word2vec vectors are trained from the contextual relationship of k-mers in the genomes, the method may extract valuable structural information from the sequences and reflect the relationship among them properly. The proposed method is optimized on the parameters from word2vec training and verified in the phylogenetic analysis of large whole genomes, including coronavirus and bacterial genomes. The results demonstrate the effectiveness of the method on phylogenetic tree construction and species clustering. The method running speed is much faster than that of the MSA method, especially the phylogenetic relationships constructed by the kmer2vec method are more accurate than the conventional k-mer alignment-free method. Therefore, this approach can provide new perspectives for phylogeny and evolution and make it possible to analyze large genomes. In addition, we discuss special parameterization in the k-mer word2vec embedding construction. An effective tool for rapid SARS-CoV-2 typing can also be derived when combining kmer2vec with clustering methods.

Structural investigations of proteins encoded by SARS-CoV-2.

It is hard to overestimate the influence of the COVID-19 pandemic on scientific research in the last two and a half years. Within a few weeks after the first cases of the disease were reported, the causative agent, now known as SARS-CoV-2, was identified, its genome was sequenced, individual proteins were expressed and purified, and structural work commenced. The originally described SARS-CoV-2 isolate (GenBank: MN908947.3) has a positive-sense single-stranded (ss) RNA genome consisting of 29,903 bases. The genome encodes 29 proteins falling into structural and nonstructural categories, expressed as polyproteins that have to be cleaved into the final products by two virally encoded cysteine proteases. This "In the Limelight" special issue of FEBS Open Bio includes three review articles, focused on different aspects of the structure and other properties of selected examples of SARS-CoV-2 proteins: (a) the properties of the Nsp14 and Nsp15 ribonucleases; (b) the current state of knowledge of the molecular mechanisms for the translation of both viral transcripts and cellular messenger RNAs, with a focus on the properties of the Nsp1 protein; and (c) the structural properties and evolution of the spike proteins in SARS-CoV-2 and other coronaviruses. These three reviews describe very different aspects of work that ultimately should lead to the development of more vaccines, antibodies, and small molecule drugs, necessary to combat this pandemic, as well as to counter future variants of this coronavirus.

Dynamics of a national Omicron SARS-CoV-2 epidemic during January 2022 in England.

Rapid transmission of the SARS-CoV-2 Omicron variant has led to record-breaking case incidence rates around the world. Since May 2020, the REal-time Assessment of Community Transmission-1 (REACT-1) study tracked the spread of SARS-CoV-2 infection in England through RT-PCR of self-administered throat and nose swabs from randomly-selected participants aged 5 years and over. In January 2022, we found an overall weighted prevalence of 4.41% (n = 102,174), three-fold higher than in November to December 2021; we sequenced 2,374 (99.2%) Omicron infections (19 BA.2), and only 19 (0.79%) Delta, with a growth rate advantage for BA.2 compared to BA.1 or BA.1.1. Prevalence was decreasing overall (reproduction number R = 0.95, 95% credible interval [CrI], 0.93, 0.97), but increasing in children aged 5 to 17 years (R = 1.13, 95% CrI, 1.09, 1.18). In England during January 2022, we observed unprecedented levels of SARS-CoV-2 infection, especially among children, driven by almost complete replacement of Delta by Omicron.

Independent acquisition of short insertions at the RIR1 site in the spike N-terminal domain of the SARS-CoV-2 BA.2 lineage.

Although the major SARS-CoV-2 omicron lineages share over 30 non-synonymous substitutions in the spike glycoprotein, they show several unique mutations that were acquired after their ancestral split. One of the most intriguing mutations associated with BA.1 is the presence of the inserted tripeptide Glu-Pro-Glu within the N-terminal domain, at a site that had previously independently acquired short insertions in several other SARS-CoV-2 lineages. Although the functional implications of the small nucleotide sequences found at this insertion hotspot, named RIR1, are still unclear, we have previously hypothesized that they may play a compensatory role in counterbalancing minor fitness deficits associated with other co-occurring spike non-synonymous mutations. Here, we show that similar insertion events have independently occurred at RIR1 at least 20 times in early 2022 within the BA.2 lineage, being occasionally associated with significant community transmission. One of these omicron sublineages, characterized by a Ser-Gly-Arg insertion in position 212, has been responsible for over 4000 documented COVID-19 cases worldwide between January and July 2022, for the most part concentrated in Denmark, where it reached a national prevalence close to 4% (10% in the Nordjylland region) in mid-May. Although the concurrent spread of the BA.2.12.1, BA.4 and BA.5 lineages led to the rapid decline of this BA.2 sublineage, the independent acquisition of several other RIR1 insertions on a BA.2 genomic background suggests that these events may provide a slight fitness advantage. Therefore, they should be carefully monitored in the upcoming months in other emerging omicron-related lineages, including BA.5.

Length-dependent motions of SARS-CoV-2 frameshifting RNA pseudoknot and alternative conformations suggest avenues for frameshifting suppression.

The SARS-CoV-2 frameshifting element (FSE), a highly conserved mRNA region required for correct translation of viral polyproteins, defines an excellent therapeutic target against Covid-19. As discovered by our prior graph-theory analysis with SHAPE experiments, the FSE adopts a heterogeneous, length-dependent conformational landscape consisting of an assumed 3-stem H-type pseudoknot (graph motif 3_6), and two alternative motifs (3_3 and 3_5). Here, for the first time, we build and simulate, by microsecond molecular dynamics, 30 models for all three motifs plus motif-stabilizing mutants at different lengths. Our 3_6 pseudoknot systems, which agree with experimental structures, reveal interconvertible L and linear conformations likely related to ribosomal pausing and frameshifting. The 3_6 mutant inhibits this transformation and could hamper frameshifting. Our 3_3 systems exhibit length-dependent stem interactions that point to a potential transition pathway connecting the three motifs during ribosomal elongation. Together, our observations provide new insights into frameshifting mechanisms and anti-viral strategies.

Cap-independent translation and a precisely located RNA sequence enable SARS-CoV-2 to control host translation and escape anti-viral response.

Translation of SARS-CoV-2-encoded mRNAs by the host ribosomes is essential for its propagation. Following infection, the early expressed viral protein NSP1 binds the ribosome, represses translation, and induces mRNA degradation, while the host elicits an anti-viral response. The mechanisms enabling viral mRNAs to escape this multifaceted repression remain obscure. Here we show that expression of NSP1 leads to destabilization of multi-exon cellular mRNAs, while intron-less transcripts, such as viral mRNAs and anti-viral interferon genes, remain relatively stable. We identified a conserved and precisely located cap-proximal RNA element devoid of guanosines that confers resistance to NSP1-mediated translation inhibition. Importantly, the primary sequence rather than the secondary structure is critical for protection. We further show that the genomic 5'UTR of SARS-CoV-2 drives cap-independent translation and promotes expression of NSP1 in an eIF4E-independent and Torin1-resistant manner. Upon expression, NSP1 further enhances cap-independent translation. However, the sub-genomic 5'UTRs are highly sensitive to eIF4E availability, rendering viral propagation partially sensitive to Torin1. We conclude that the combined NSP1-mediated degradation of spliced mRNAs and translation inhibition of single-exon genes, along with the unique features present in the viral 5'UTRs, ensure robust expression of viral mRNAs. These features can be exploited as potential therapeutic targets.

Description of a One-Year Succession of Variants of Interest and Concern of SARS-CoV-2 in Venezuela.

Some of the lineages of SARS-CoV-2, the new coronavirus responsible for COVID-19, exhibit higher transmissibility or partial resistance to antibody-mediated neutralization and were designated by WHO as Variants of Interests (VOIs) or Concern (VOCs). The aim of this study was to monitor the dissemination of VOIs and VOCs in Venezuela from March 2021 to February 2022. A 614 nt genomic fragment was sequenced for the detection of some relevant mutations of these variants. Their presence was confirmed by complete genome sequencing, with a correlation higher than 99% between both methodologies. After the introduction of the Gamma VOC since the beginning of the year 2021, the variants Alpha VOC and Lambda VOI were detected as early as March 2021, at a very low frequency. In contrast, the Mu VOI, detected in May 2021, was able to circulate throughout the country. After the detection of the Delta VOC in June 2021, it became the predominant circulating variant. With the arrival of the Omicron VOC in December, this variant was able to displace the Delta one in less than one month.

The snoGloBe interaction predictor reveals a broad spectrum of C/D snoRNA RNA targets.

Box C/D small nucleolar RNAs (snoRNAs) are a conserved class of RNA known for their role in guiding ribosomal RNA 2'-O-ribose methylation. Recently, C/D snoRNAs were also implicated in regulating the expression of non-ribosomal genes through different modes of binding. Large scale RNA-RNA interaction datasets detect many snoRNAs binding messenger RNA, but are limited by specific experimental conditions. To enable a more comprehensive study of C/D snoRNA interactions, we created snoGloBe, a human C/D snoRNA interaction predictor based on a gradient boosting classifier. SnoGloBe considers the target type, position and sequence of the interactions, enabling it to outperform existing predictors. Interestingly, for specific snoRNAs, snoGloBe identifies strong enrichment of interactions near gene expression regulatory elements including splice sites. Abundance and splicing of predicted targets were altered upon the knockdown of their associated snoRNA. Strikingly, the predicted snoRNA interactions often overlap with the binding sites of functionally related RNA binding proteins, reinforcing their role in gene expression regulation. SnoGloBe is also an excellent tool for discovering viral RNA targets, as shown by its capacity to identify snoRNAs targeting the heavily methylated SARS-CoV-2 RNA. Overall, snoGloBe is capable of identifying experimentally validated binding sites and predicting novel sites with shared regulatory function.