Wednesday, July 21, 2021

One of these things doesn't belong: efforts to exclude problematic sequences in GRCh38

Since the release of GRCh38, the GRC has received a number of user reports alerting us to a potential false duplication involving chr 21p and 21q. Users noted that reads were aligning to both regions in GRCh38, but not GRCh37/hg19, resulting in a decreased mapping score and difficulties in variant calling throughout. Additionally, user analyses involving Multiplex Ligation-dependent Probe Amplification (MLPA), a technique for gene copy number detection, and exome studies indicated potential false duplications. The implicated regions contained several genes, including CBS (Gene ID: 875), U2AF1 (Gene ID: 7307) and KCNE1B (Gene ID: 3753). The GRC has investigated the matter and concurs that the GRCh38 assembly contains sequence on the short arm of chr 21 that should be excluded from analyses. Read on to learn more about this issue, as well as some recently detected non-human contamination in GRCh38, and ways you can find and avoid these sequences in your analyses.

The short arm of human chromosome 21, like that of the four other human acrocentric chromosomes, is where genes associated with rDNA synthesis are localized, and is characterized by highly repetitive heterochromatic sequence. The repetitive nature of these sequences, coupled with limitations in sequencing technology, have until recently made the representation of these regions in genome assemblies very difficult. 

As a consequence, the GRCh37 representation of the chromosome 21 p-arm contained only 11 clone sequences. Seven were clones from the HSA21-specific BAC library CHORI-507 that had previously been experimentally localized to 21p (PMID: 17895424). In an effort to add additional sequence to this repetitive region, 23 additional components were added to 21p for GRCh38, including 18 additional CHORI-507 clones, 4 RPCI-11 clones, and 1 ABC9 fosmid. Admixture mapping localized some of these clones to this region.

In response to the user reports, the GRC re-reviewed the sequences added to 21p in GRCh38. Haploid CHM13hTERT Illumina reads generated by The McDonnell Genome Institute were aligned to GRCh38 by NCBI, and evaluated for read mapping and coverage. This analysis supported the user reports, suggesting that 5 of the newly added CHORI-507 clones (FP565260.4, CU639417.17, FP236240.8, FP475955.4 and CU633980.13) were actually redundant with sequences on chr 21q, and thus represented false duplications in GRCh38.

The GRC has now removed these sequences from the files that it uses to generate the reference assembly. However, we cannot remove them from the GRCh38 assembly without triggering the next major release of the human assembly. In order to help users recognize these regions and avoid them in their analyses, we have produced a masking file to be used as a companion to GRCh38. This BED file is available from the GenBank FTP site: This file provides the assembly coordinates of the 5 clones incorrectly localized to chr 21p. The Genome in a Bottle Consortium recently posted a preprint demonstrating that using this masking file greatly improves variant calling accuracy in the affected genes (

In addition to these sequences, the file also includes 2 other assembly scaffolds that were found, after the release of GRCh38, to be contaminated with non-human sequence. These include a chr Un scaffold (KI270752.1/NT_187507.1), whose sole component (AF065393.1) is now known to represent sequence from Chinese hamster (PMID:30486838), likely derived from the human-hamster CHO cell line that was the clone source, and an alternate loci scaffold (KI270825.1/NT_187580.1) whose non-anchor component (AC225822.3) was shown to be chimeric. In AC225822.3, the first 25,375 bases are human sequence matching GRCh38 chr 10 reference component and alternate scaffold anchor sequence AL391421.27, while the rest match Acidithiobacillus thiooxidans sequences from multiple WGS projects (PMID:32398145). Although all sequences in the reference assembly are screened for foreign contamination, these two were not detected at the time of release (2014). Prompted by these findings, the GRC has more recently re-screened the assembly with updated contamination databases and has not detected additional issues. As these two scaffolds are not human sequence, very few reads are likely to map well to them, but users may still want to make note of them in their analyses.

In total, the contamination represents ~800 Kb, or 0.02% of the total sequence length. GRCh38 remains an extremely high quality reference assembly. Nonetheless, the GRC remains committed to addressing assembly errors and making sure it serves as the most reliable analysis substrate possible. Check out our website to see other genomic regions under review. We welcome your feedback and reports of newly discovered issues!  In the future, we plan to update the masking file with any new regions as identified and reviewed by the GRC.

 Fig. A

Fig. A:  Aligned CHM13hTERT Illumina reads viewed in Integrative Genomics Viewer (IGV). The panes labeled 'Original' were reads aligned prior to redundant sequence masking and the panes labeled 'Fixed' are reads aligned after redundant sequence masking.  

The top two panes show reads aligned to the valid U2AF1 locus in GRCh38 (NC_000021.9:43,091,000-43,110,000 of 21q) and the bottom two panes show reads aligned to the falsely duplicated (pseudo region) region of GRCh38 (NC_000021.9:6,480,000-6,500,000 of 21p). 

In IGV, sequence reads that align to 2 places in the Reference (whether it is correct or not), yield poor/ambiguous alignments, indicated by clear, unshaded reads. This is shown by both the 'Original hg38 U2AF1' and the 'Original hg38 pseudo region' pane. 

Following the masking of the known, duplicated region introduced in GRCh38, the aligned reads in the 'Fixed hg38 U2AF1 gene' pane are shaded grey, meaning they have good mapping scores to that region. And there are no reads mapping to the 'Fixed hg38 pseudo region' because the duplicated sequence is masked in the Fixed hg38 file.

Fig. B

Fig. B: Aligned reads to 21q region that has false duplication on 21p in GRCh38 before masking. Note the BAC clone boundary where alignment of falsely duplicated region in 21p starts. This duplication involves the CBS gene (Gene ID: 875).

Fig. C

Fig. C: Aligned reads to 21p region falsely duplicated (in Fig. B). You can see ambiguous read alignment and the falsely duplicated CBS gene annotated in the gene track.

Fig. D

Fig. D: Alignment of BAC FP236240.8 (redundant BAC added to 21p for GRCh38) to the corresponding valid region on 21q. Note the redundant BAC alignment to the region (bottom pane) and the valid read alignment depth (shown in middle pane). Since the region was falsely duplicated, read alignment in the region of redundant BAC alignment is poor.

Monday, November 30, 2020

A New Rat Genome Assembly Sparks Membership of Rat and RGD in the Genome Research Consortium

The Rat Genome Database RGD is very pleased to announce the release of mRatBN7.1, the new rat genome assembly! The mRatBN7 assembly, generated by the Darwin Tree of Life Project at the Wellcome Sanger Institute, is significantly improved over the Rnor6.0 and previous assemblies. mRatBN7 was derived from a male BN/NHsdMcwi rat that is a direct descendent from the female BN rat previously sequenced. The new BN rat reference genome was generated using multiple technologies including PacBio long reads, 10X linked reads, Bionano maps and Arima Hi-C. Its quality is a substantial improvement compared to any of the previous assemblies, with just 175 scaffolds and having an N50 >135Mb and 756 contigs with N50 >29Mb, resulting in a contiguity similar to the human or mouse reference assemblies. The assembly has been submitted to the International Nucleotide Sequence Database Consortium (INSDC), and the initial GenBank record for it is now available at Genome annotation, i.e. the assignment of gene positions and prediction of new genes and other genomic elements, will be generated by both NCBI and Ensembl. We are also pleased to announce that Rat and the mRatBN7 assembly have been accepted into the Genome Reference Consortium (GRC) and the RGD has been approved to represent the rat research community and participate in the ongoing work of curating the assembly. RGD will work closely with curators from the GRC, with the International Rat Omics Consortium (IROC), a grassroots community of rat genomics researchers, and the rat research community to identify any candidate regions for focused genome curation. Stay tuned for the appearance of rat on the GRC website!
Hi-C 2D map of mRatBN7.1 generated with HiGlass

Wednesday, July 22, 2020

GRCm39: the new mouse reference genome assembly

The GRC is pleased to announce the release of GRCm39 (GCA_000001635.9), the latest version of the mouse reference genome assembly. 

GRCm39 is the first coordinate-changing update to the mouse reference since the 2012 release of GRCm38. More than 400 reported issues were resolved in the production of the new assembly, which also incorporates the sequence edits released as scaffolds in the six GRCm38 patch releases.

The new reference assembly exhibits substantial improvements in contiguity. As shown in Fig 1, the scaffold N50 has increased by 95% to 106.1 Mb in GRCm39, and 1.9 Mb of non-N bases were added to the assembly. The gap count has been nearly cut in half, with the total gap length reduced by 4.5 Mb. The decrease in gap length reflects in part the use of optical map data to size the remaining gaps wherever possible, replacing many of the default 50 kb gaps found in GRCm38. Sequences used for gap closures included clones, GRC-constructed contigs, as well as contigs from the C57BL/6J long-read based assembly ASM377452v2.

GRCm39 assembly statistics
Figure 1: GRCm39 Assembly Statistics

As in prior assembly versions, the GRCm39 chromosome sequences continue to represent the C57BL/6J strain. However, the alternate loci scaffolds that provided additional strain representations for highly variant genomic regions in GRCm38 and MGSCv37, have been removed from the assembly. The relatively low usage of these scaffolds, coupled with a growing number of high quality strain-specific genome assemblies available in public sequence databases, such as those generated by the Mouse Genomes Project, has reduced the need for the inclusion of these sequences in the reference genome assembly. Although no longer affiliated with the reference assembly, these sequences remain available in the INSDC databases (GenBank/ENA/DDBJ).

The new reference assembly will be annotated by GENCODE and RefSeq in the coming months. An in-depth transcript alignment analysis of a pre-release version of the GRCm39 assembly, presented at the 2019 IMGC meeting, demonstrated that there is improved representation for more than 50 genes. A list of these genes is provided in our earlier blog post. The GRC will provide a complete list of genes improved in GRCm39 as the annotation effort progresses.

Notable curation activities represented in the new assembly, but not in previous patch releases, include the targeted update of more than 1,500 individual bases at which the GRCm38 allele representation was erroneous or an unsupported C57BL/6J variant, a substantial retiling of the chr X pseudo-autosomal region (PAR) that provides representation for several genes missing from GRCm38 (Fig 2), removal of a false triplication involving the Duxbl locus, and correction of a 16 Mb inversion at the proximal end of chromosome 14.

GRCm39 chromosome X
Figure 2. Genes in GRCm39 chr X PAR

The GRC wishes to thank the many members of the mouse community who have reported assembly issues, and contributed their time, expertise, and data to assist in curation efforts. Updates to the GRC website will be made to reflect the new assembly. With the release of GRCm39, the GRC's curation of the mouse genome reference assembly will be limited to the resolution of community reported problems. We encourage you to contact the GRC for additional information on the curation of assembly regions of interest. You can also subscribe to grc-announce email list to receive email notification for all GRC assembly updates.

Monday, June 8, 2020

ZFIN and the GRC: Supporting the zebrafish reference genome assembly

ZFIN is a member of The Genome Reference Consortium (GRC), an international collaboration consisting of NCBI, the Wellcome Trust Sanger Institute, the McDonnell Genome Institute at Washington University, the European Bioinformatics Institute (EBI) and ZFIN. This consortium is tasked with ensuring that the reference assemblies for human, mouse, zebrafish and chicken are updated and improved through new data and analysis from genome centers and the research community.

The zebrafish-specific GRC webpage (Fig. 1) provides an overview of the zebrafish genome, including an ideogram of the latest zebrafish assembly (GRCz11) that highlights the location of alternate loci scaffolds, downloadable files for the current public assemby and the tiling path files reflecting the latest assembly edits, as well as links to genome assembly data and genome regions under review. Zebrafish genome issues (Fig. 2), such as sequencing errors, gaps, and path problems, can be browsed at the chromosome level, filtered by problem type or status or searched by gene, location, clone name or accession.

If you come across what you suspect is a problem in the build in the course of your research, visit the GRC website to search the list of genome issues and if it has yet to be reported, select the "Report an Issue" tab in the header (Fig. 3) to report information about the potential problem in the build. Be as complete as possible and provide location, flanking sequences and a description of the issue. Genome annotators will evaluate the region, determine if an update to the genome is needed and submit data to create a new tiling path to improve the build with an update or "patch".

We welcome your feedback!

Figure 1 Zebrafish specific GRC webpage
Figure 1

Figure 2 Zebrafish genome issues webpage
Figure 2

Figure 3 Report a Genome Problem webpage
Figure 3

Thursday, May 23, 2019

Readying the release of GRCm39

GRCm38, the current mouse reference assembly, whose chromosomes represent the C57BL/6J strain, supports a broad range of research activities. Despite being one of the highest quality mammalian genome assemblies ever produced, it still has more than 600 gaps and includes sub-optimal representations for some genes. To address these issues and provide the murine research community with an improved substrate for their work, the GRC has been applying new technologies, such as optical/genome mapping, and using new sequence resources to curate an update to the reference genome assembly. The public release of the updated assembly, GRCm39, is planned for the end of 2019/early 2020.

Since the 2012 release of GRCm38, the last coordinate-changing update to the mouse reference, the GRC has provided 6 publicly accessible minor assembly updates, the last of which (GRCm38.p6) was released in September, 2017. These non-coordinate changing assembly versions, known as patch releases, cumulatively include 65 fix patches (chromosome path changes) and 9 novel patches (alternate representations of chromosome sequences, derived from other strains). In GRCm39, these fix patches will be incorporated into the chromosomes, and the novel patches will persist as alternate loci.

In preparation for the release of GRCm39, the GRC also analyzed a non-public updated version of the reference assembly. In the production of this updated assembly, 322 reported genome issues were resolved, 70% of which addressed gaps and problems with the sequence of underlying genomic clone components. The review evaluated the value and impact of the released patches and subsequent unreleased genome updates and assessed the need for additional work prior to the GRCm39 release.

Analyses of this assembly, known informally as "GRCm38B", reveal the removal of about 200 Kb over-expanded sequence found in GRCm38, and the addition of 254 new components, of which 95 are contigs assembled from WGS reads (PRJNA51977). Due to the curation effort, GRCm38B has fewer gaps, and increased contig and scaffold N50s in comparison to GRCm38 (Table 1).

Furthermore, an analysis of mouse RefSeq transcripts aligned to GRCm38B demonstrates the improved representation for at least 50 genes (Table 2).

One such example of improved gene representation in GRCm38B is shown in Figure 1. In GRCm38, an assembly gap at chromosome 4 nt 99,842,111, between components BX324127.8 and CU326395.5, results in a partial representation of Efcab7 (EF-hand calcium binding domain 7). In GRCm38B, sequences from MF597759.1 (a GRC-assembled contig of Illumina reads) closes this gap and provides the exons missing from Efcab7 transcript NM_145549.1.
Figure 1 Top: Incomplete representation of Efcab7 gene in GRCm38 due to an assembly gap. Middle: The gap is closed in GRCm38B with MF597759.1. It also provides complete representation of Efcab7. Bottom: Efcab7-mRNA partial representation in GRCm38 and complete representation in GRCm38B.

Based on GRCm38B analyses, 5 GRCm39 chromosomes will be comprised of a single scaffold (Chr. 11, 12, 15, 16, 18), 11 will be built from 2 scaffolds and the remaining 5 from more than 2 scaffolds.

In the months leading up to the GRCm39 release, the GRC will continue to curate additional genome issues. Sequences from the recently published C57BL/6J long-read based assembly ASM377452v2 are providing new resources for the update or closure of assembly gaps and correction of sequencing errors. Additionally, we are investigating individual bases at which the GRCm38 sequence differs from all 17 strain-specific genome assemblies (Mouse Genome Project) with the aim of correcting confirmed erroneous bases.

Upon the release of GRCm39, the GRC's curation of the mouse genome reference assembly will be limited to the resolution of community reported problems.

You can browse the status of GRC curation activities at our website, and we encourage you to contact the GRC for additional information on the curation of assembly regions of interest. Updates to the timeline for the GRCm39 release will be provided on the Mouse Genome Overview webpage. You can also subscribe to grc-announce email list to receive email notification for all GRC assembly updates .

Tuesday, March 26, 2019

Shining a light on human acrocentric p-arms

The GRC is excited to announce that representations for the p-arms of the human acrocentric chromosomes can now be found in the GRCh38.p13 patch update of the reference genome, thanks to work done in Brian McStay's lab. These sequences are included on the following scaffolds: ML143366.1, ML143367.1, ML143372.1, ML143377.1, and ML143380.1.

The p-arms of the human acrocentric chromosomes HSA13-15, 21 and 22 each bear ribosomal gene arrays (Figure 1) termed nucleolar organiser regions (NORs). These are the most transcriptionally active regions of the genome and direct formation of nucleoli, the largest structures in the nuclei of all human cells. Research on these critical genomic regions is hampered by the fact that acrocentric p-arms are not included in human genome drafts. They are both internally highly repetitive and share a strikingly similar sequence content, making them recalcitrant to standard sequencing approaches. Despite these issues, Brian McStay's lab previously described a collection of sequenced cosmid and BAC clones that allowed them build a reasonable consensus for sequences both immediately proximal and distal to NORs (Floutsakou et al. 2013. Genome Res 23:2003-12). Proximal sequences are almost entirely segmentally duplicated, similar to regions bordering centromeres. In contrast, the distal sequence is predominantly unique to the acrocentric p-arms. Their interphase localisation, open chromatin structure and transcriptionally active state, point to a role in nucleolar biology and prioritise their inclusion in a future genome draft (for discussion see McStay. 2016 Genes Dev. 30:1598-610).

The McStay lab subsequently developed a workflow that has enabled them to determine the NOR distal sequence, the Distal Junction (DJ) from all five acrocentric chromosomes and from an additional two versions of HSA21, ~3 Mb in all. A panel of mono-chromosomal somatic cell hybrids, mouse A9 cells containing individual human chromosomes, allowed them to sequence one chromosome at a time. Sequencing was performed by combining sequence capture with PacBio SMRT sequencing. Pre-capture libraries (typically in the range of 4-6 kb) were prepared from each hybrid line. Capture was performed using oligonucleotide libraries designed using their original consensus. Circular consensus sequencing (CCS) of post-capture libraries generated so called reads of insert (ROIs) each with high sequence accuracy. This allowed the McStay group to assemble sequence contigs from the NOR distal region of each chromosome, regardless of the presence of repetitive sequences such as satellite DNA.

Their analysis of these sequences confirms sequence and presumably functional conservation between the acrocentrics. It also provides evidence for non-homologous exchanges between them. It's anticipated that extension of sequence contigs towards the telomeres will uncover increased structural variation between the acrocentric chromosomes.

Figure 1. FISH experiment showing the relative locations of the rDNA array and distal junctions on the p-arms of the human acrocentric chromosomes.

Wednesday, March 20, 2019

GRCh38.p13 has been released

The GRC is pleased to announce that GRCh38.p13 is now available! This release adds 45 new scaffolds: 43 FIX patches and 2 NOVEL patches. The FIX patch scaffolds provide assembly corrections while the NOVEL patch scaffolds deliver new alternate sequence representations. A valuable contribution to this patch release comes in the addition of the Nucleolus Organiser Region (NOR) sequences for the short arms of the acrocentric chromosomes (13, 14, 15, 21, and 22) as provided by Brian McStay's group (PMID: 23990606). The NOR additions will be discussed in detail in a separate blog.

With access to an ever-increasing pool of high quality, long-read human assembly data, the GRC has been able to utilise this in GRCh38.p13 to address genome issues that have until now persisted due either to lack of data, or complexity. Much of the data added in this patch is derived from the CHM1 human haploid hydatidiform mole assembly (GCA_001297185.2). Originally produced as part of an assembly comparison analysis (see PMID: 
28396521), the assembly was recently Pilon corrected and re-submitted to GenBank by the McDonnell Genome Institute at the Washington University, a GRC center, with the specific aim of improving the base pair accuracy for use of its sequences in improving the Human Genome Reference.

In GRCh38.p13, a total of 28 assembly gaps have been closed. These updates, together with sequences added to correct 5 clone errors, add more than 0.5 Mb of unique data to the assembly. The majority of unique sequences added in this release come from contigs that are components of WGS assemblies derived from PacBio sequence reads, such as the CHM1 assembly mentioned above. However, genomic clone libraries still play an important role in assembly curation. In this release BAC clones from human cell line (CHM1htert) have provided complete, single haplotype representations of clinically important regions such as Prader-Willi on chromosome 15, and CT47 on chromosome X.

The CT47 cancer/testis antigen located on human Xq24 is organized as an array of 4.8 kb tandemly repeated units. Due to the repetitive nature of the sequence involved, coupled with the limitations of the technologies available at the time, the representation of the CT47 gene cluster in GRCh37 and GRCh38 was problematic. The region is gapped, and the flanking clones are from different haplotypes. As a consequence, the representation of the cluster in these assemblies was incomplete and biologically unsound, representing an indeterminate number of gene copies (Figure 1, top).
Studies have indicated that this polymorphic array is highly variable between haplotypes and ranges from 4 to 17 copies in length. Long-read sequencing of genomic clones has now captured the complete CT47 cluster as a single haplotype. The fix patch (ML143381.1) included in the GRCh38.p13 release now provides a contiguous and validated representation of the CT47 genomic region. This patch closes the assembly gap with sequences from BAC clone AC275592.1 (CH17-182I12) which contains a complete, 7 copy representation of the CT47 array (Figure 1, bottom). Note that this update reduces the number of CT47 genes represented as compared to GRCh37 and GRCh38.
Figure 1 Top: CT47 region in GRCh38. Incomplete representation of CT47 gene cluster in GRCh38 due to an assembly gap. Bottom: CT47 fix patch in GRCh38.p13. The gap is closed and a complete representation of CT47 cluster is provided. 
Optical mapping technology has been used to confirm the copy number for the CT47 array is accurate for the haploid CHM1tert sample (Figure 2), from which the clone library was derived.
Figure 2: AC275592.1 alignment to CHM1 Bionano optical map.

As more data becomes available using the latest technologies the GRC is able to utilise this in order to continually to update and improve the reference genome. If you have questions about this process, let us know.

You can download the GRCh38.p13 assembly, including the alignments of the patches to GRCh38, from the GenBank FTP.