Category Archives: Transhuman News

A new systematic review of the validated monogenic causes of male infertility strengthens the evidence base for emerging gene-disease relationships; the review hopes to encourage more routine genetic testing in clinics and identify gaps in our knowledge of male infertility genetics.

In less than two years, the number of genes revealed in the literature as causative of male infertility phenotypes has increased from 78 (in 2019) to 104, an increase of 33%. All are supported by evidence of a direct gene-disease relationship and, say the authors of a new report, will provide the impetus for an update of existing guidelines, will inform novel evidence-based genetic testing strategies used in clinics, and will identify gaps in our knowledge of male infertility genetics.(1)

This latest report is an update on the first standardised clinical validity assessment of monogenic causes of male infertility published in 2019, and like that one this latest literature evaluation has been conducted with the International Male Infertility Genomics Consortium (IMIGC). The increase in the number of high-probability male infertility genes is not just a reflection of extended gene discovery but a function too of the recent rapid uptake of next generation sequencing in male infertility (with whole-exome sequencing described as the default sequencing approach) and research on clinical cohorts. The validation of these emerging genes will, say the authors, help give direction to which individual genes may be screened for and how they are relevant to certain types of infertility.

All 104 genes and their links to male infertility phenotypes are listed in a table, with organ effects noted in hypothalamic function, pituitary and adrenal gland dysfunction, vas deferens, reproductive organ development, Leydig cell, Sertoli-cell only syndrome, meiotic arrest, spermatogenesis and fertilisation.

As illustrated in a Campus meeting on the genetics of male infertility held online a few weeks ago, a wide range of phenotypes now appears to be largely genetic in origin.(2) Already, the genetic components of Klinefelter syndrome, Y-chromosome microdeletions and some monogenic causes of azoospermia are well recognised, but there remains a majority of male infertility cases (60-70%) without any clear diagnosis. Only 4% are actually diagnosed with a defined genetic cause. This updated systematic evaluation of all available evidence for published monogenic causes of isolated or syndrome male infertility will hopefully extend the limits for genetic testing and the diagnostic power for identifying the causes of male infertility; however, while NGS is now a cornerstone test in male infertility research, it is not, say the authors, extensively employed in clinical diagnosis.

The report notes that the diagnostic rate of genetic tests for all types of isolated male infertility currently lies between 4 and 9%. These are rates considerably behind those seen in other heterogeneous disorders with a large genetic contribution such as cardiomyopathies or developmental delay, where whole-exome and whole-genome sequencing are routinely applied. However, with reduced costs of NGS and better accessibility, the authors hope that the increased number of validated genes implicated in male infertility will lead to greater diagnostic implementation.

And this in turn will help in evaluating future health risks in which male infertility may be linked to other comorbidities in later life; for example, a number of multiple DNA repair genes linked to infertility are known to be involved in some cancers. There are also implications from genetic testing in male infertility for sons of men with Y-chromosome variants conceived by ART, who presumably will inherit the same infertility phenotype as their affected fathers - and thus may in turn require ART if wanting their own biological child.

1. Houston BJ, Riera-Escamilla A, Wyrwoll MJ, et al. A systematic review of the validated monogenic causes of human male infertility: 2020 update and a discussion of emerging genedisease relationships, Hum Reprod Update 2021; doi:10.1093/humupd/dmab0302. See https://www.focusonreproduction.eu/article/ESHRE-News-male

See the original post:
The number of genotype-phenotype associations validated in male infertility continues to grow - ESHRE

A decline in the use of the word "race" in papers on human genetics reflects a growing understanding of race as a social construct. But other trends may point to ongoing uncertainty about how to discuss different populations.

What to know:

Human geneticists have moved away from using the word "race" to describe populations, a study recently published in The American Journal of Human Genetics (AJHG) shows.

Researchers examined the text of all 11,635 articles published between 1949 and 2018 by the AJHG. While the word "race" appeared in 22% of papers in the first 10 years of the paper's publication, it was used in just 5% of papers in the last 10 years.

This decline points to the current understanding in science of race as a social construct and a desire to move away from past research that erroneously conflated genetics with racial categories, according to lead author Vence Bonham, JD, the acting deputy director of the National Human Genome Research Institute.

The study also found that the alternative and sometimes more ambiguous terms "ethnicity" and "ancestry" have increased over time, which may suggest that geneticists are still struggling to find terms to accurately describe populations.

A The National Academies of Science, Engineering, and Medicine has recently formed a committee to produce a consensus report on the use of the word "race" and other terms descriptive of populations in health disparities research.

This is a summary of the article "Human geneticists curb use of the term 'race' in their papers" published by Science on December 2. The full article can be found on science.org.

For more news, follow Medscape on Facebook, Twitter, Instagram, YouTube, andLinkedIn

See the original post:
Geneticists Have Reduced Use of the Term 'Race' in Papers - Medscape

Flash isnt your average puppy. A yellow Labrador, named after one of the first British guide dogs from 1931, she is playful, affectionate, and loves learning new commands. Flash is enrolled in an elaborate program herself, one that takes two years and nearly $50,000 to train her to become a guide dog for the blind and visually impaired. Her temporary caregiver Melanie will make sure she maintains a healthy routine: twice-daily walks in different environments, a train ride here, a trip to a mall there to get used to other people. But Melanie has already accomplished one of her most important tasks: When Flash was five months old, she swabbed the puppys cheek and mailed the saliva away to a team of researchers that is trying to decipher the link between dog genetics, health, and behavior.

Around half of the dogs that are bred for guiding dont end up doing that work because of health or behavioral problems. Modern dogs suffer from many genetic diseases, a side effect of keeping breeds separate and selecting them for desirable traits. Some of these purebreds might have the right looks, but not the right temperament, to become a working dog. But what if breeders could predict what makes a good guide dog and select against undesired traits, ensuring they arent passed on to the next generation?

More than 500 traits analogous to human genetic conditions have been described in dogsboth species can suffer from cancer, eye disease, or dysplasia of the hip, to name a few. Cheap DNA tests for canines can screen for changes, known as mutations, in a single gene. The causes of many other conditions, however, are more complex. They can be linked to multiple genes or to environmental factors like exercise, food, dust, or mold spores. We definitely want to get a handle on complex traits, says Tom Lewis, head of canine genetics at Guide Dogs. The charity breeds around 1,000 puppies a year, which spend their first year in the homes of volunteers before entering formal training.

Before joining Guide Dogs in January, Lewis worked at the Animal Health Trust and the Kennel Club in the United Kingdom, where he studied the genetic risk of hip dysplasia in breeds registered with the club. Dysplasia is one of the hereditary conditions that can be difficult to diagnose and treat. It is a malformation of the hip joint that develops during growth, though traumatic injury, being overweight, or lacking muscle strength can worsen the condition. For example, puppies raised in homes with hardwood flooring may build less muscle mass in their legsthey cant get traction on the floor, and slip and slide around, which is hard on their little joints. The constant pain can eventually turn into lameness and arthritis in grown dogs, making them unsuitable for guiding or assisting people with disabilities.

Good health is key for guide dogs, but temperament is just as important. They need to lead their owners around obstacles and other people while staying calm and obedient. They need to resist chasing after squirrels or getting too excited when meeting other dogs. Not every breed has what it takes. For example, the typical cocker spaniel is intelligent, affectionate, and a great option for families, but it is also too excitable. Even if you give them the same training, you would never expect a spaniel to be a guide dog. They're far too temperamentally unsuited, and that's probably a genetic thing, says Lewis.

The rest is here:
Is There a Genetic Link to Being an Extremely Good Boy? - WIRED

A gene variant initially uncovered in the genomes of people belonging to the Old Order Amish has been linked in a new study to lower levels of fibrinogen (a blood clotting factor) and low-density lipoprotein (LDL) cholesterolboth of which, when elevated, increase a persons risk of developing cardiovascular disease. The work, published today (December 2) in Science,not only connects a missense mutation in the enzyme-coding gene beta-1,4-galactosyltransferase 1 (B4GALT1) with heart health in humans, but confirms the link in mice.

This is a very good example of the utility of small founder or isolated populations in predicting genetic effects of genes that could not easily be identified even in the very big human biobanks that are available worldwide, Caroline Hayward, who studies human genetics at the University of Edinburgh and did not participate in the study, writes in an email to The Scientist.

May Montasser, a genetic epidemiologist at the University of Maryland School of Medicine, and colleagues study the genomes of the Old Order Amish because approximately 35,000 people alive today can trace their family history back to a small number of founder families. Due to the small pool of genetic starting material, this population harbors less genetic diversity than the general population, meaning that variants that might disappear in larger groups can be maintained in Old Order Amish populations and therefore be easier to spot.

In the new study, Montasser and colleagues sequenced the exons of 6,890 Old Order Amish subjects and found a missense mutation in B4GALT1,which is expressed throughout the body and encodes an enzyme that transfers galactose to proteins. The variant was associated with lower LDL cholesterol and was present in six percent of the Amish genomes but is much rarer outside the Amish community. The researchers found it in only eight of 140,000 non-Amish genomes that are part of a National Heart, Lung, and Blood Institute database.

When the team looked at other factors related to cardiovascular health in people carrying the missense mutation, they found no association with triglycerides and a small association with high density lipoprotein cholesterol. The blood clotting factor fibrinogen, which can be a risk factor for cardiovascular disease when elevated, was lower in people with the variant.

To assess whether the missense mutation in B4GALTwas linked with overall cardiovascular health, the researchers shifted their focus outside the relatively healthy Amish population to the Geisinger Health System and the UK Biobank, two larger genomic databases. Because the specific variant they identified in the Amish is so rare in the general population, Montasser and her colleagues pulled out genomes with any variant of B4GALT.These individuals had decreased LDL, fibrinogen, and a 35 percent reduction in cardiovascular disease.

When the researchers generated knock-in mice with the B4GALTvariant, the animals also had lower LDL and fibrinogen. Knocking the gene down just in the rodents livers led to lower levels of circulating LDL. The authors explain in the paper that this finding could point to the usefulness of targeting B4GALT expression in the liver therapeutically to lower LDL cholesterol.

The study provides strong evidence that this newly discovered mutation is relevant across population and species, Kari North, a genetic epidemiologist at the University of North Carolina at Chapel Hill who did not participate in the work, writes in an email to The Scientist. B4GALT may represent a new drug target for decreasing LDL-cholesterol and downstream [cardiovascular disease]. However, years of work are still needed to develop this new discovery into a new pharmaceutical target, she adds.

Before getting to the point of clinical relevance, we have to make sure that there are not any harmful side effects [associated with] having this variant, Montasser explains. Having low LDL is great; having low fibrinogen is great, but is there anything else harmful? So far, based on all the other information we have, everything looks good. Those people look perfectly healthy, but we are doing even more deep phenotyping right now to make sure that we are not missing anything.

Another issue is that its not clear how we go from having this variant to having low LDL and low fibrinogen and protection from cardiovascular disease, she says. The research team is trying to characterize that mechanism in animal models and human samples now. Montasser says they will keep working on it [and] hopefully someday well have some form of therapy based on this.

See the original post:
Genetic Variant Discovered in Amish Protects from Heart Disease - The Scientist

Abstract

In less than 25 y, the field of animal genome science has transformed from a discipline seeking its first glimpses into genome sequences across the Tree of Life to a global enterprise with ambitions to sequence genomes for all of Earths eukaryotic diversity [H. A. Lewin etal., Proc. Natl. Acad. Sci. U.S.A. 115, 43254333 (2018)]. As the field rapidly moves forward, it is important to take stock of the progress that has been made to best inform the disciplines future. In this Perspective, we provide a contemporary, quantitative overview of animal genome sequencing. We identified the best available genome assemblies in GenBank, the worlds most extensive genetic database, for 3,278 unique animal species across 24 phyla. We assessed taxonomic representation, assembly quality, and annotation status for major clades. We show that while tremendous taxonomic progress has occurred, stark disparities in genomic representation exist, highlighted by a systemic overrepresentation of vertebrates and underrepresentation of arthropods. In terms of assembly quality, long-read sequencing has dramatically improved contiguity, whereas gene annotations are available for just 34.3% of taxa. Furthermore, we show that animal genome science has diversified in recent years with an ever-expanding pool of researchers participating. However, the field still appears to be dominated by institutions in the Global North, which have been listed as the submitting institution for 77% of all assemblies. We conclude by offering recommendations for improving genomic resource availability and research value while also broadening global representation.

The first animal genome sequence was published 23 y ago (1). The 97 millionbasepair (bp) (Mb) Caenorhabditis elegans genome assembly ushered in a new era of animal genome biology where genetic patterns and processes could be investigated at genome scales. As genome assemblies have accumulated for an increasingly diverse set of species, so too has our knowledge of how genomes vary and shape Earths biodiversity (e.g., refs. 2 and 3). Major shifts in genome availability and quality have been driven by two key events. First, the invention of high-throughput, short-read sequencing provided an economical means to generate millions of reads for any species from which sufficient DNA could be obtained. These 100-bp short reads could be assembled into useful, albeit fragmented, genome assemblies. Later, the rise of long-read sequencing allowed for similarly economical generation of reads that are commonly orders of magnitude longer than short reads, resulting in vastly more contiguous genome assemblies (4).

We have now entered an era of genomic natural history. Building on 250 y of natural history efforts to describe and classify the morphological diversity of life on Earth, we are gaining a complementary genomic perspective of Earths biodiversity. However, a baseline accounting of our progress toward a complete perspective of Earths genomic natural historywhere every species has a corresponding, reference-quality genome assembly availablehas not been presented. This knowledge gap is particularly important given the momentum toward sequencing all animal genomes, which is being driven by a host of sequencing consortia. For instance, the Vertebrate Genomes Project seeks to generate high-quality assemblies for all vertebrates (5), the Bird10K project seeks to generate assemblies for all extant birds (6), the i5K project plans to produce 5,000 arthropod genome assemblies (7), the Earth BioGenome Project aims to sequence all eukaryote genomes (8), and the Darwin Tree of Life project plans to sequence genomes for all eukaryotes in Britain and Ireland (https://www.darwintreeoflife.org/).

In this Perspective, we curated, quantified, and summarized genomic progress for a major component of Earths biodiversity: kingdom Animalia (Metazoa) and its roughly 1.66 million described species (9). We show that as of June 2021, 3,278 unique animals have had their nuclear genome sequenced and the assembly made publicly available in the National Center for Biotechnology Information (NCBI) GenBank database (10). This translates to 0.2% of all animal species. When viewed through the lens of major clades, massive disparities exist. For instance, 32 times more assemblies are available for chordates than arthropods (Fig. 1).

Variation in taxonomic richness and genome availability, quality, and assembly size across kingdom Animalia in GenBank (as of 28 June 2021). Taxonomic groups are clustered by phylogeny following ref. 11. Only groups with 30 or more available assemblies as of January 2021 are shown with the exception of Hominidae (n = 5 assemblies). In the tree, bold group names represent phyla and naming conventions follow those of the NCBI database. Of 34 recognized animal phyla, 10 do not have a representative genome sequence. (A) The total number of described species for each group following Zhang (9) and the references therein. (B) Genomic representation among animal groups for 3,278 species with available genome assemblies. Bars represent the magnitude of the observed minus the expected number of genomes given the proportion that each group comprises of described animal diversity. Significance was assessed with Fishers exact tests and significantly under- or overrepresented groups (P < 0.05) are denoted with asterisks. Gray numbers indicate the total number of species with available genome assemblies for each group. The number of available assemblies is not mutually exclusive with taxonomy; that is, a carnivore genome assembly would be counted in three categories (order Carnivora, class Mammalia, phylum Chordata). (C) The percentage of described species within a group with an available genome sequence (bars) and the percentage of those assemblies that have corresponding annotations (red circles). For many groups (e.g., arthropods), only a fraction of a percent of all species have an available genome assembly, making their percentage appear near zero. (D) Assembly size for all animal genome assemblies, grouped by taxonomy. (E) Contig N50 by taxonomic group. The sequencing technology used for each assembly is denoted by circle fill color: short-read (blue), long-read (yellow), or not provided (gray). In D and E, each circle represents one genome assembly and a few notable or outlier taxa are indicated with gray text.

To construct a database of the best available genome assembly for all animals, we downloaded metadata from GenBank for all kingdom Animalia taxa using the summary genome function in v.10.9.0 of the NCBI Datasets command-line tool on 4 February 2021. Next, we used the TaxonKit (12) lineage function to retrieve taxonomic information for each taxid included in the genome metadata. To gather additional data for each assembly (e.g., sequencing technology), we used a custom web scraper script. Both this web scraper script and the scripts used to download and organize the metadata are available in this studys GitHub repository (https://github.com/pbfrandsen/metazoa_assemblies). We later supplemented this initial dataset with a second round of metadata acquisition on 28 June 2021. For the full dataset, we hand-refined the NCBI taxonomy classifications to subdivide our dataset into three categories: species, subspecies, or hybrids (Dataset S1). If replicate assemblies for a taxon were present, we defined the best available assembly as the one with the highest contig N50 (the midpoint of the contig distribution where 50% of the genome is assembled into contigs of a given length or longer).

We filtered our data in several ways: We removed subspecies (unless they were the only representative for a species), hybrids, and assemblies that were shorter than 15.3 Mb [the smallest confirmed assembly size for a metazoan to date (13)] or had a contig N50 less than 1 kilobase (Kb). We also culled assemblies that were unusually short (i.e., 1 to 2.5 Mb) with information in their descriptions that indicated they were not true nuclear genome assemblies (e.g., exon capture). In total, we culled 407 assemblies based on the above criteria. The remaining assemblies were classified as short-read, long-read, or not provided if only short reads (e.g., Illumina) were used, any long-read sequences (e.g., PacBio) were used, or no information was available. We defined a species as having gene annotations available if any assembly for that taxon also had annotations in GenBank. When the best available assembly did not have annotations included or when multiple assemblies had annotations, we retained the annotations for the assembly with the highest contig N50. Finally, we used the submitting institution for each assembly as a surrogate for the institution that led the genome assembly effort. Using these data, we classified assemblies to a country, region (Africa, Asia, Europe, Middle East, North America, Oceania, South America, Southeast Asia), and the Global North (e.g., Australia, Canada, Europe, United States) or Global South (e.g., Africa, Asia including China, Mexico, Middle East, South America).

To test if clades were under- or overrepresented in terms of genome availability relative to their species richness, we compared the observed number of species with assemblies with the expected total for the group. We obtained totals for the number of described species overall and for each group from previous studies, primarily from Zhang (9) and the references therein. We assessed significance between observed and expected representation with Fishers exact tests (alpha = 0.05). We tested for differences in distributions of contig N50 or assembly size between short- and long-read genomes with Welchs t tests. For both display (i.e., Fig. 1) and analysis, we subdivided the dataset into the lowest taxonomic level that still contained 30 or more assemblies as of January 2021 (with the exception of hominids, which were given their own category due to their exceptionally high genomic resource quality).

Genome assemblies were available for 3,278 species representing 24 phyla, 64 classes, and 258 orders (Fig. 2A and Dataset S1). The dataset was exceptionally enriched for the phylum Chordata (which includes all vertebrates) with 1,770 assemblies for the group (54% of all assemblies) despite chordates comprising just 3.9% of animal species (P, Fishers < 1e-5; Fig. 1). Conversely, arthropods were underrepresented with 1,115 assemblies (34% of the dataset) for a group that comprises 78.5% of animal species (P, Fishers < 1e-5; Fig. 1). However, not all arthropods were underrepresented; five insect clades were overrepresented (Apidae [bees], Culicidae [mosquitoes], Drosophila [fruit flies], Formicidae [ants], and Lepidoptera [butterflies and moths]; all P, Fishers < 1e-3; Fig. 1). Collectively, of the 59 animal taxonomic groups included in our dataset, 14 groups were underrepresented, 17 were represented as expected, and 28 were overrepresented (primarily chordates; Fig. 1). Ten phyla had no publicly available genome sequence (Fig. 1). Over the 17-y GenBank genome assembly record, animal assemblies have been deposited at a rate of 0.52 species assemblies per day. Over the most recent year, however, this rate increased eightfold to 4.07 assemblies per day. If the most recent rate were maintained, all currently described animals would have a genome assembly available by 3136. To achieve this goal by 2031 instead, an average of 165,614 novel animal genomes would need to be sequenced and assembled each year (112 times faster than the rate for the most recent year).

Genome availability for kingdom Animalia versus taxonomic descriptions and over time. (A) The proportion of described taxonomic groups versus the number with sequenced genome assemblies from phyla to species. The gray plot (Right) is a zoomed-in perspective of the higher taxonomy-level categories in the full plot (Left). For genus through phylum, the number of described categories is based on the NCBI taxonomy. For species, the total number described is from Zhang (9). (B) The timeline of genome contiguity versus availability for animals according to the GenBank publication date (x axis; C). A rise in assembly contiguity has been precipitated by long-read sequencing. Particularly contiguous assemblies for a given time period are labeled. (C) The number of animal genome assemblies deposited in GenBank each month since February 2004. Several notable events are labeled. When specific dates are indicated, those (and the assemblies referred to) are included within that months total. For B and C, it is important to note that when a genome assembly is updated to a newer version, its associated date is also updated. Thus, the date associated with many early animal assemblies [e.g., C. elegans (1)] has shifted to be more recent with updates.

The average animal genome assembly was 1.02 gigabases (Gb) in length (SD 1.21 Gb) with a contig N50 of 2.26 Mb (SD 25.16 Mb; Fig. 1 D and E). Two animal genome assemblies were 25 Gb longer than all other assembliesthe axolotl [32.4 Gb (14)] and Australian lungfish [34.6 Gb (15)] (Fig. 1D). The smallest genome assembly in the dataset, the mite Aculops lycopersici, was over 1,000 times smaller, spanning just 32.5 Mb (16). Still smaller is the 15.3 Mb assembly of the marine parasite Intoshia variabili, which has the smallest animal genome currently known (13). But, since the I. variabili assembly was not available in GenBank as of June 2021, it was not included in our dataset.

Contiguity varied dramatically across groups. For instance, hominid assemblies (family Hominidae, n = 5) were the most contiguous with an average contig N50 of 24.2 Mb. Bird assemblies (class Aves, n = 515) were also highly contiguous (mean contig N50 = 1.4 Mb) despite being so numerous (and accumulating over a long period of time). On the other end of the spectrum, jellyfish and related species (phylum Cnidaria) exhibited some of the least contiguous genome assemblies with a mean contig N50 of 0.18 Mb (n = 65; Fig. 1E). Roughly 34% of animals with genome assemblies had corresponding annotations in GenBank but annotation rates differed substantially among groups (Fig. 1C). For example, the rate of arthropod annotations (22.3%) lags behind that for chordates (41.3%); however, much of this disparity appeared to be driven by the low and high annotation rates of butterflies and moths (order Lepidoptera) and birds (class Aves), respectively. Of 445 assemblies, just 6.5% of lepidopteran assemblies in GenBank have corresponding annotations versus 72.8% of birds (n = 519 assemblies; Fig. 1C). Notably, since most gene models are based on sequence similarity to known functional genes and not functional data, the true rate of annotation is likely even lower than reported here.

Animal genome assemblies have been contributed by researchers at institutions on every continent with permanent inhabitants, including 52 countries. From a regional perspective, institutions in North America (n = 1,331), Europe (n = 972), and Asia (n = 828) collectively accounted for 95.5% of all assemblies (Fig. 3A). And, nearly 70% of all animal genome assemblies have been submitted by researchers in just three countries: United States (n = 1,275), China (n = 676), and Switzerland (n = 317) (Fig. 3A). When countries were grouped by their inclusion in the Global North or South, similarly stark patterns emerged. Researchers affiliated with institutions in the Global North contributed roughly 75% of animal genome assemblies (Fig. 3B). From a taxonomic perspective, researchers at North American institutions have contributed the most insect and mammal assemblies, European researchers have contributed the most fish assemblies, and Asian researchers have contributed the most bird assemblies (Fig. 3A). The first assembly in GenBank from the Global North was deposited in 2004 and the first assembly from the Global South was deposited in 2011 (Fig. 3C). Since then, the number of assemblies deposited each year has steadily risen, with the proportions from the Global North and South staying relatively constant (Fig. 3C).

Where animal genome assemblies have been produced around the world according to the submitting institutions in GenBank. (A) For each geographic region, total numbers of genome assemblies are shown by dark circles with white lettering. This total is further broken down by country and taxon. For regions where more than four countries have contributed assemblies (e.g., Europe), an Other category represents all other countries. The same applies to all assemblies that are not insects, birds, fish, or mammals in the taxon plots. Countries are color-coded by assignment to the Global North or South. (B) The total number of genome assemblies contributed by countries in the Global North (e.g., United States, Europe, Australia) versus the Global South (e.g., Africa, South America, China, Mexico, Middle East). (C) The rate of genome assembly deposition by major sources in the Global North (Europe, United States) and Global South (China, Southeast [SE] Asia) as well as all other countries collectively in each (Other).

Use of long reads in genome assemblies and availability of key metadata also differ with geography. For assemblies deposited since 2018, researchers from the Global South have used long reads slightly more frequently than those from the Global North (25.7% versus 20.2%; Fig. 4A). However, researchers from the Global North were far less likely to report the types of sequence data used (19.9% of assemblies for the Global North versus 1.4% of assemblies for the Global South; Fig. 4A). Much of this difference appears to be driven by genome assemblies deposited by researchers at European institutions (Fig. 4B). This gap in metadata may reflect an issue with data mirroring between the European Nucleotide Archive (ENA) and GenBank. For instance, many new genome assemblies being generated by the United Kingdom, for example, are part of the Wellcome Sanger Institutes Darwin Tree of Life project, which is generating exceptionally high quality assemblies using long-read sequencing and depositing them into the ENA (Fig. 5). One region (Oceania) and three countries (Australia, Finland, India) reported long reads being used in more than 50% of deposited assemblies (Fig. 4 B and C).

Sequencing technologies used around the world (A) between the Global North versus Global South, (B) among regions, and (C) among countries. To limit bias due to the limited availability of long-read sequencing technologies before 2017 (Fig. 2B), only assemblies deposited on or after 1 January 2018 were included in the analysis and in C only countries that deposited five or more assemblies during the focal period (January 2018 to June 2021) are shown.

Examples of major contributors of genome assemblies for (A) butterflies (order Lepidoptera), (B) birds (class Aves), and (C) fish (primarily class Actinopterygii). Major contributors were defined as any consortium, organization, or project that has deposited more than 5% of all assemblies for butterflies and birds or 2.5% of all assemblies for fish.

Animal genome sequencing has dramatically progressed in the last 25 y. In that span, the field has moved from sequencing the first nuclear genome for any animal (1)a landmark achievementto targeting the generation of genome assemblies for all of Earths eukaryotic biodiversity (8). Here, we provided a contemporary perspective on progress toward this goal for the 1.6 million species in the animal kingdom (9). We showed that while tremendous progress has been made, major gaps and biases remain both in terms of taxonomic and geographic representation, at least within the most commonly used database of genomic resources, GenBank. For instance, a major bias exists in favor of vertebrates which are vastly overrepresented relative to their total species diversity (Fig. 1 AC). From the perspectives of biomedicine and human evolution, this bias is reasonable since humans are vertebrates. However, from a basic research perspective, particularly as it relates to genomic natural history and an overarching goal to sequence all animal genomes, there is a need to taxonomically diversify sequencing efforts.

At the highest taxonomic levels, 10 animal phyla still have no genomic representation. To illustrate the scale of this disparity versus other groups and the unique biology that is being overlooked, genome assemblies are available for 685 ray-finned fishes (class Actinopterygii) but none exists for phylum Nematomorpha, an 2,000-species clade of parasitic worms whose presence can dramatically alter energy budgets of entire stream ecosystems (17). Another phylum without genomic representationLoriciferawas first described in 1983 (18). This group of small, sediment-dwelling animals includes the only examples of multicellular species that spend their entire life cycles under permanently anoxic conditions (19). Loriciferans accomplish this feat by foregoing the energy-producing mitochondria found in virtually all animals in favor of hydrogenosome-like organelles akin to those found in prokaryotes inhabiting anaerobic habitats (19). Clearly, there is much to discover in terms of genomic diversity and functional biology in clades yet to be sampled.

A few select countriesprimarily the United States, several European nations, and Chinahave led the sequencing of the vast majority of animal genome assemblies (Fig. 3A). Aside from China, all of these countries are within the Global North. This pattern of geographic bias raises two potential issues for representation in animal genome science. First, the researcher population of animal genome sequencing likely does not reflect the global population. Second, sampling biases may exist toward the regions where most of the genome sequencing is occurring. Some of this bias is intentional and reflects funding goals for a given region. For instance, the Darwin Tree of Life project seeks to sequence the genomes of all 70,000 eukaryotic species living in Britain and Ireland. Still, however, similar to how sampling biases can yield skewed understanding of the natural world in other disciplines (e.g., ref. 20), so too could bias toward specific ecoregions, habitats, or other classifications skew genomic insight.

Inherently linked with questions of representation in animal genome science is the specter of parachute science (or helicopter research)the practice where international scientists, typically from wealthy nations, conduct studies in other countries that are often poorer without meaningful communication nor collaborations with local people (21). Parachute science has a long history in ecological research, and signatures of these practices have been observed for genome sciences. For instance, Marks etal. (22) found that the majority of plant genome assemblies for species that are native to South America and Africa were sequenced off-continent by researchers at European, North American, or Asian institutions. Given the sheer number of animal genome assemblies that have been submitted by a small number of countries and institutions, a similar pattern likely exists for animal genomes. However, to properly assess this issue, parsing authorship to quantify collaboration, at a minimum, would need to occur and this approach would still overlook key aspects of representation that need to be considered (e.g., if a researcher from the Global South is working at an institution in the Global North).

For the purpose of biological discovery, not all genome assemblies are created equal. As long-read sequencing technologies have matured, so too has the quality of assemblies being generated (4). In the last year alone, the largest ever animal genome assembly was deposited [Australian lungfish (15)] as well as the most complete human genome to date, a telomere-to-telomere assembly (23). Still, many species in GenBank only have low-quality assemblies available (i.e., contig N50 < 100 Kb with no corresponding gene annotations; Fig. 1). Since fragmentation and/or poor or missing gene annotations reduce the research value of an assembly, genome quality is important, particularly when the end goal is resource development for a broader community. As of April 2021, the Earth BioGenome Project sought assembly quality of 6.C.Q40 (https://www.earthbiogenome.org/assembly-standards) for reference genomes, where 6 refers to a 1e-6 contig N50 (i.e., 1 Mb). In our dataset, 568 assemblies (17.3%) reach this contiguity standard. And that number drops to 271 assemblies (8.3%) when contig N50 1 Mb and deposited gene annotations are both required. For reference, the C above refers to chromosomal scale scaffolding and Q40 to a less than 1/10,000 error rate. Neither of these metrics were assessed in this study.

Independent research laboratories, institutions, and consortia have contributed genome assemblies on both ends of the quality spectrum (Fig. 5). For example, among butterflies (order Lepidoptera), a bimodal quality distribution is being primarily driven by contributions made in 2021 by two submitting institutions, the Florida Museum of Natural History (e.g., ref. 24) and the Wellcome Sanger Institute (Fig. 5A). When viewing genome assembly contributions holistically across the animal Tree of Life, it is clear that two consortiathe Vertebrate Genomes Project (5) and the Darwin Tree of Life, part of the Wellcome Sanger Institutewarrant specific recognition for contributing exceptional genomic resources relative to closely related species (Fig. 5).

While animal genome science has dramatically matured in recent years, the field still rests on the cusp of massive change. Thousands of genome assemblies are now available for a wide range of taxa, a resource that can empower unprecedented scales of genomic comparison. Simultaneously, multiple consortia are building momentum toward their goals and generating some of the highest-quality genome assemblies ever produced. The field is also diversifying, with researchers around the world, particularly from the Global South, leading a rising number of efforts. These ongoing advances will yield higher-quality, more globally representative genome data for animals. As we collectively build toward this new genomic future, we offer recommendations to improve assembly quality and accessibility while also continuing to increase representation within the discipline.

The quality of a genome assembly is likely the most important factor dictating its long-term value. Genome assembly quality, however, is difficult to define. Here, we propose a holistic view on genome assembly quality that generally echoes the guidelines proposed by the Earth BioGenome Project and other consortia. Briefly, assemblies should reach minimum levels of contiguity (e.g., contig N50 > 1 Mb) and accuracy in order to be considered a reference that will likely not need to be updated for most applications. At a minimum, assemblies should also include high-quality gene annotations that perhaps take advantage of standardized pipelines [e.g., NCBI Eukaryotic Genome Annotation Pipeline (25)] to maximize compatibility across taxa. We recommend the field further improve the quality of genome assembly resources in two ways. First, refining and expanding the coordinated deposition of genome assemblies will improve the usability of the resources and reproducibility of analyses. It will also reduce duplications of effortthat is, when a group sequences a genome that has already been producedan issue that is likely to become increasingly common.

To refine and expand coordinated resource deposition, we recommend the continued use of GenBank (10) or one of the other archives that are members of the International Nucleotide Sequence Database Collaborationthe ENA and DNA Database of Japanas the central repositories for genome assemblies and their metadata given their tripartite data-sharing agreement. Next, we call on genetic archive administrators, consortia, and independent researchers to collectively improve the metadata submitted with each assembly and the mirroring of data across repositories. Too many assemblies lack basic information about the sequence data and methods used (e.g., Fig. 4) and, with the difficulty of linking assemblies to published studies (if available), it can be challenging or impossible to find this information. Further, an expansion of the metadata associated with each assemblyideally to make more of the categories required and expand demographic datawould make efforts to quantify geographic representation, for instance, far more straightforward. Alternatively, the metadata associated with genome assembly accessions could be integrated with existing efforts like the Genomic Observatories Metadatabase [GeOMe (26)]. Furthermore, a set of minimum quality characteristics for a genome assembly may need to be defined. A number of exceptionally low quality genome assemblies (e.g., with contig N50 values shorter than 1 Kb) that often cover only a small fraction of the expected total genome sequence length for a given group are present in GenBank. The presence of these assemblies raises the question: Where is the inflection point between resource quality and value to other researchers versus diluting the resources of a shared repository?

For our second recommendation, we amplify and expand the message of Buckner etal. (27) and Thompson etal. (28): Genome science needs specimen vouchers. Vouchers serve as a key physical link between taxonomy and molecular insight. Rarely, however, are vouchers referenced in publications of genome assemblies; only 11% of vertebrate assemblies included such a reference as of January 2020 (27). While vouchers represent a physical reference for assessing taxonomic classification or morphological variation, a properly stored voucher could also provide a long-term source of material for future resource improvement. If a physical specimen cannot be deposited, photographs and/or genomic DNA should be deposited in its place (e.g., ref. 29). Tied to the metadata discussion above, additional fields should be added to GenBank genome assembly accessions to directly link the assembly to a specimen, photo, or genomic DNA that has been deposited elsewhere.

Though geographic representation in animal genome science has improved in recent years, the discipline appears far from properly reflecting the global researcher pool. This issue is almost certainly multifaceted, likely stemming from a lack of infrastructure (e.g., fewer high-throughput sequencing platforms in developing countries), fewer resources for expensive molecular research, and a corresponding lack of training in genome data analysis. To bridge this gap and to empower a more diverse discipline, the nations and institutions that are devoting large amounts of resources to animal genome sequencing (e.g., China, United Kingdom, United States), and the researchers within those countries, should continue to develop meaningful collaborations with researchers within countries where their focal species reside (30). These meaningful collaborationswhere all parties are valued for their expertise and involved in decision makingimprove the science through transfer of local knowledge, provide a means for local researchers to expand their skillset and network while raising their scholarly profile, and, most importantly, can effectively end the practice of parachute science (30). Within-continent (or -country) initiatives also have transformative potential for people and genome research. For instance, the African-led effort to sequence 3 million African genomes over the next 10 y (the 3MAG project) will yield massive investment in African genomics, an incredible resource for understanding the full scope of human genetic diversity, and a new generation of African genome scientists (31). While focused on human genetics, the infrastructure and expertise that arise from the 3MAG project will no doubt translate to other taxa in the coming years.

A practical justification also exists for increasing representation in genome science, particularly as we seek to generate genome assemblies for every animal on Earth. The Global South is home to the bulk of the worlds biodiversity (32) and, as such, researchers in these regions have greater access to key habitats and specimens. Thus, it behooves everyone, including researchers in the Global North, to deepen collaborations with peers in the Global South while also helping to build indigenous capacity for collection, storage, and sequencing of new specimens.

Animal genome science continues to grow and expand at an exceptional rate. The coming years will surely see thousands, and perhaps tens of thousands, of new genome assemblies from across the Tree of Life, technological and analytical improvements, and some of the largest-scale and most in-depth studies of animal genome biology conducted to date. However, if we are to realize the ambitious goals of efforts like the Earth BioGenome Projecta self-described biological moonshotthe rate and mean quality of animal genome assembly production will have to increase by roughly two orders of magnitude. Regardless of rates and timelines, however, perhaps the most important goal for the future of animal genome science is that we empower a more diverse, representative researcher community in parallel with the generation of new resources.

All study data are included in the article and/or supporting information.

S.H. and J.L.K. were supported by NSF Award OPP-1906015. We thank Guangfeng Song, Eric Cox, and Anne Ketter from the Datasets development team at the NCBI for their responsiveness and receptiveness to improving this valuable tool for data science.

Author contributions: S.H., J.L.K., and P.B.F. designed research; S.H. and P.B.F. performed research; S.H. and P.B.F. analyzed data; and S.H., J.L.K., and P.B.F. wrote the paper.

The authors declare no competing interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2109019118/-/DCSupplemental.

Read the rest here:
Toward a genome sequence for every animal: Where are we now? - pnas.org