The Ensembl COVID-19 resource: Ongoing integration of public SARS-CoV-2 data

ABSTRACTThe Ensembl COVID-19 browser (covid-19.ensembl.org) was launched in May 2020 in response to the ongoing pandemic. It is Ensembl’s contribution to the global efforts to develop treatments, diagnostics and vaccines for COVID-19, and it supports research into the genomic epidemiology and evolution of the SARS-CoV-2 virus. This freely available resource incorporates a new Ensembl gene set, multiple sets of variants, and alignments of annotation from several resources against the reference assembly for SARS-CoV-2. It represents the first virus to be encompassed within the Ensembl platform. Additional data are being continually integrated via our new rapid release protocols alongside tools such as the Ensembl Variant Effect Predictor. Here we describe the data and infrastructure behind the resource and discuss future work.

Lost in Translation: The Pitfalls of Ensembl Gene Annotations Between Human Genome Assemblies and Their Impact on Diagnostics

Background:The GRCh37 human genome assembly is still widely used in genomics despite the fact an updated human genome assembly (GRCh38) has been available for many years. A particular issue with relevant ramifications for clinical genetics currently is the case of the GRCh37 Ensembl gene annotations which has been archived, and thus not updated, since 2013. These Ensembl GRCh37 gene annotations are just as ubiquitous as the former assembly and are the default gene models used and preferred by the majority of genomic projects internationally. In this study, we highlight the issue of genes with discrepant annotations, that have been recognized as protein coding in the new but not the old assembly. These genes are ignored by all genomic resources that still rely on the archived and outdated gene annotations. Moreover, the majority if not all of these discrepant genes (DGs) are automatically discarded and ignored by all variant prioritization tools that rely on the GRCh37 Ensembl gene annotations.Methods:We performed bioinformatics analysis identifying Ensembl genes with discrepant annotations between the two most recent human genome assemblies, hg37, hg38, respectively. Clinical and phenotype gene curations have been obtained and compared for this gene set. Furthermore, matching RefSeq transcripts have also been collated and analyzed. ٌResults:We found hundreds of genes (N=267) that were reclassified as “protein-coding” in the new hg38 assembly. Notably, 169 of these genes also had a discrepant HGNC gene symbol between the two assemblies.Most genes had RefSeq matches (N=199/267) including all the genes with defined phenotypes in Ensembl genes GRCh38 assembly (N=10). However, many protein-coding genes remain missing from the current known RefSeq gene models (N=68)Conclusion: We found many clinically relevant genes in this group of neglected genes and we anticipate that many more will be found relevant in the future. For these genes, the inaccurate label of “non-protein-coding” hinders the possibility of identifying any causal sequence variants that overlap them. In addition, Important additional annotations such as evolutionary constraint metrics are also not calculated for these genes for the same reason, further relegating them into oblivion.

Lost in translation: the pitfalls of Ensembl Gene annotations between human genome assemblies and their impact on diagnostics

BackgroundThe GRCh37 human genome assembly is still widely used in genomics despite the fact an updated human genome assembly (GRCh38) has been available for many years. A particular issue with relevant ramifications for clinical genetics currently is the case of the GRCh37 Ensembl gene annotations which has been archived, and thus not updated, since 2013. These Ensembl GRCh37 gene annotations are just as ubiquitous as the former assembly and are the default gene models used and preferred by the majority of genomic projects internationally. In this study, we highlight the issue of genes with discrepant annotations, that have been recognized as protein coding in the new but not the old assembly. These genes are ignored by all genomic resources that still rely on the archived and outdated gene annotations. Moreover, the majority if not all of these discrepant genes (DGs) are automatically discarded and ignored by all variant prioritization tools that rely on the GRCh37 Ensembl gene annotations.MethodsWe performed bioinformatics analysis identifying Ensembl genes with discrepant annotations between the two most recent human genome assemblies, hg37, hg38, respectively. Clinical and phenotype gene curations have been obtained and compared for this gene set. Furthermore, matching RefSeq transcripts have also been collated and analyzed.ResultsWe found hundreds of genes (N=267) that were reclassified as “protein-coding” in the new hg38 assembly. Notably, 169 of these genes also had a discrepant HGNC gene symbol between the two assemblies. Most genes had RefSeq matches (N=199/267) including all the genes with defined phenotypes in Ensembl genes GRCh38 assembly (N=10). However, many protein-coding genes remain missing from the current known RefSeq gene models (N=68)ConclusionWe found many clinically relevant genes in this group of neglected genes and we anticipate that many more will be found relevant in the future. For these genes, the inaccurate label of “non-protein-coding” hinders the possibility of identifying any causal sequence variants that overlap them. In addition, Important additional annotations such as evolutionary constraint metrics are also not calculated for these genes for the same reason, further relegating them into oblivion.

The Prognostic Power of Gene Expression Profiling with Cytogentics and Routinely Acquired Serum Markers: SKY92 Combined with Revised ISS

Background Updating prognostic models for multiple myeloma is important in the context of changing treatment options. Previously we have described the value of the prognostic marker SKY92, which identifies high-risk multiple myeloma patients, as well as the value of the combined SKY92-ISS marker. With the introduction of revised ISS, it is of interest to evaluate the value of the updated combination of SKY92 with R-ISS. Within the HOVON87/NMSG18 trial, stratification into 3 groups was described: high-risk: 11% SKY92 high-risk (HR) + R-ISS II-III, low-risk: 15% SKY92 standard risk (SR) + R-ISS I and intermediate risk (74%, other). The 3-year PFS rates were 54% (95%CI: 38-77%), 27% (95%CI: 21-37%) and 7% (95%CI: 1-46%) for SKY-RISS I, II and III, respectively (p < 0.001). The 3-yr OS rates for SKY-RISS I to III were 88%, 66% and 26% (p=6×10-7). Here we describe the validation of SKY92-RISS in the CoMMpass dataset. Methods SKY92 was determined using RNA-seq data available from the CoMMpass dataset. Briefly, the SKY92 score was obtained as a weighted summation of the expression given by the available Ensembl gene IDs, corresponding to the probe sets of the SKY92 classifier. Renormalization of the original SKY92 discovery data (HOVON65/GMMG-HD4) allowing a direct remodeling between the Affymetrix probe-set expressions (i.e. SKY92) and RNAseq Ensembl gene IDs. Only Ensembl gene IDs with an average log2 expression >8 were used. Revised ISS status was determined as described. For optimal comparison to the discovery cohort of the HOVON87/NMSG18 trial, the analysis was limited to 93 patients older than 65 years in the CoMMpass data set, that did not receive transplant, and for whom RNA-Seq at diagnosis, R-ISS and follow-up data were available. Results The median follow-up is 41 months. SKY92 identified 24 high-risk patients (24/93: 26%). The 3-yr PFS and OS rates of standard-risk patients were 49% and 80% respectively, compared to 23% and 44% for high-risk, resulting in a significant log rank test (p < 0.005). The R-ISS classified patients into the low-risk R-ISS I (24% of patients), intermediate-risk R-ISS II (63%) and high-risk R-ISS III (13%). The 3-yr PFS rates were 76% (RISS I), 33% (RISS II) and 33% (RISS III); for OS: 100% (RISS I), 68% (RISS II) and 33% (RISS III; PFS, p = 0.07; OS, p < 0.001). SKY92 and R-ISS were independent prognostic factors in terms of OS and PFS. The SKY-RISS classification resulted in 20% low-, 61% intermediate- and 18% high-risk patients (Figure 1). The 3-yr PFS rates were 81% (95%CI: 64-100%), 42% (95%CI: 30-59%) and 12% (95%CI: 3-44%; p < 0.001) and 3-yr OS rates were 100% (95%CI: 100-100%), 77% (95%CI: 66-89%) and 32% (95%CI: 16-61%; p <0.001). Out of 69 patients classed as standard risk using the SKY92 classifier (80% 3-yr OS rate), 17 and 52 were classified as SKY-RISS I and II, respectively, resulting in a 3-yr survival rate of 100% and 74%, respectively. In contrast, out of 24 SKY92 HR patients (44% 3 yr OS rate), 5 were classified as SKY-RISS II (100% alive at 3 years) with the remainder true high-risk patients (32% alive at 3 years). Out of 12 RISS III patients (3-yr OS, 33%), 5 were classified as SKY-RISS II (3-yr OS: 60%) and 7 as SKY-RISS III (3-yr OS: 14%). Conclusion This study demonstrates the value of gene expression profiling - SKY92 - alongside revised ISS. They form a solid combination, improving on either marker separately. Both models combined clearly identified more high-risk patients correctly, whilst also placing low risk patients into a more appropriate risk category. This was shown in the discovery set and was subsequently applied to an independent set, confirming the validity and usability of the SKY-RISS. Disclosures Kuiper: SkylineDx: Current Employment, Current equity holder in private company. Van Vliet:SkylineDx: Current Employment, Current equity holder in private company. Van Beers:SkylineDx: Ended employment in the past 24 months. Zweegman:Celgene: Membership on an entity's Board of Directors or advisory committees; Oncopeptides: Membership on an entity's Board of Directors or advisory committees; Sanofi: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees, Research Funding; Takeda: Membership on an entity's Board of Directors or advisory committees, Research Funding. Broijl:Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene/BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees. Sonneveld:Sanofi: Consultancy; Bristol-Myers Squibb: Consultancy, Honoraria, Research Funding; Takeda: Consultancy, Honoraria, Research Funding; Skyline Dx: Honoraria, Research Funding; Karyopharm: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria, Research Funding; Amgen: Consultancy, Honoraria, Research Funding.