iiinova — where your digital self keeps evolving

Darrin T. Schultz @dts.bsky.social · 27/08/2026

More broadly, how we cite data hasn't caught up to the scale of these datasets. Separately, I found that genome assembly databases (like NCBI) often don't have originating publications linked for a genome. Sometimes there is no paper, but many times there is a paper, it is just missing.

000

Darrin T. Schultz @dts.bsky.social · 27/08/2026

What I can do now is to keep working on recovering the originating publication for most of these assemblies from NCBI and Europe PMC, and improve table S1 on Zenodo and the relevant github repos published in this paper (chrombase, genbargo, egt).

110

Darrin T. Schultz @dts.bsky.social · 27/08/2026

That's fair. We weren't able to cite thousands of papers in the manuscript, but table S1 lists every accession and its NCBI submitter. In the manuscript we acknowledge institutions at the >20-genome level, but that under-credits the many groups contributing fewer genomes.

100

Reposted by Darrin T. Schultz

voigtvision.bsky.social @voigtvision.bsky.social · 20/08/2026

🤩 It's time for some invertebrate eye candy! 🤩 Here's a ctenophore (or comb jelly) stained with phalloidin by John Alexander Salazar Hernandez & imaged on an exaSPIM built by Kaelin Wulf of the team of @adamkglaser.bsky.social as part of the FHL Functional Biodiversity course #microscopy

213732

Darrin T. Schultz @dts.bsky.social · 20/08/2026

Thank you, Max. Would be happy to chat about it sometime.

010

Darrin T. Schultz @dts.bsky.social · 20/08/2026

Yes, it's a bit crazy! This image looks cleaner than the real "picture" of the genome because I'm comparing the animals to a predefined set of genes. When you compare their genomes directly there are more genes that have undergone dispersal to other chromosomes. Will try to remember to follow up...

140

Darrin T. Schultz @dts.bsky.social · 19/08/2026

That's all for now! Thank you for reading, thanks to our editors and reviewers for helping us improve our manuscript, and contact us/me if you have any questions!

140

Darrin T. Schultz @dts.bsky.social · 19/08/2026

I would like to note that 39% of the genomes were from labs or institutions that submitted 20 or fewer! Individual contributions are still critical, even in the age with large sequencing consortia. Contact me if you'd like me to connect you to resources to start your own genome assembly project!

The full list of genomes used in this study, as well as the percent
of genomes contributed by submitter, is in table S1. We recognize
the substantial investments made by genome sequencing consortia
and individual laboratories in producing the chromosome- scale ge-
nomes analyzed in this study. Below, we briefly note the contributors
of 0.75% or more of the total genomes to our dataset based on NCBI
“Assembly Submitter”: The Wellcome Sanger Institute (previously
known as “The Sanger Centre” and “Wellcome Trust Sanger Insti-
tute”; 2655 genomes or 45.62% of genomes in the dataset), Geno-
scope CEA (117, 2.0%), the Vertebrate Genomes Project (108, 1.86%),
Northwest A&F University (83 genomes, 1.43%), the DNA Zoo (52
genomes, 0.89%), and the US Department of Agriculture (45, 0.77%).
Additional institutions contributing more than 20 genomes include
Zhejiang University; China Agricultural University; Centro Nacio-
nal de Análisis Genómico; Institute of Oceanology, Chinese Acade-
my of Sciences; Institut de Biologia Evolutiva Consell Superior
d’Investigacions Científiques- Universitat Pompeu Fabra (CSIC- UPF);
University of Cambridge; Nanjing Agricultural University; European
Molecular Biology Laboratory’s European Bioinformatics Insti-
tute (EMBL- EBI), Wellcome Trust Genome Campus, Cambridge;
Chinese Academy of Agricultural Sciences; The Max Planck Institute
of Molecular Cell Biology and Genetics; Bat1K Project; Southwest
University; GenoFish; Ocean University of China; Princeton Uni-
versity; Rothamsted Research; and Huazhong Agricultural University.
The remaining 39.23% of genomes in the dataset were submitted to
NCBI by contributors with 20 or fewer chromosome- scale genome
uploads after filtering.

1100

Darrin T. Schultz @dts.bsky.social · 19/08/2026

We are very grateful to all of those who published the 5821 genomes, including the largest three institutional contributors @sangerinstitute.bsky.social, Genoscope CEA, and the @vertebrategenomes.bsky.social. We respected institutional embargoes in our data selection!

240

Darrin T. Schultz @dts.bsky.social · 19/08/2026

4. The smaller pieces, each standalone. odp draws chromosome-scale dot plots between any two species. chromsim simulates fusion-with-mixing. breakpointer2 infers inversion breakpoints from whole-genome alignments. Check the supplement for links to the software! github.com/conchoecia/odp

120

Darrin T. Schultz @dts.bsky.social · 19/08/2026

3.5 The second clickable is the And the orthology graph database, written by Fatih Sarıgöl at the University of Vienna, queryable in your browser: metazoa.csb.univie.ac.at/neo4j

Graph representation of orthology and genome context across species. This figure
depicts the graph database encoding the relationships between the 4,454 species with
chromosome-scale genomes in our dataset. (A) Data model used in the Neo4j database. Nodes
represent biological entities: chromosome (CHR), gene (GENE), orthogroup (OG), and ancestral
linkage group (ALG). Solid “location” edges (relationships) connect a feature (node) to the
chromosome on which it resides within a genome. “Orthology” edges connect genes to
orthogroups (OGs). Example node properties stored in the database (e.g., taxid, assembly,
species, lineage, genomic coordinates) are indicated in dashed boxes. ( B ) This subgraph shows
the genomic relationship between the ALG A1a and the genomes of the jellyfish Rhopilema
esculentum (red) and the scallop Pecten maximus (blue). For each species, there are genes
connected to (orthologous to) the BCnS ALG A1a OGs. These displayed genes are connected to
the chromosome on which they exist for both species. The Pecten and Rhopilema chromosomes
have been highly conserved since the myriazoan ancestor, and retain single chromosomes for the BCnS A1a ALG. Node size encodes entity type, as in panel A, and as shown in the legend: large
nodes are chromosomes (CHR), medium gray nodes are orthogroups (OG), small nodes are
genes (GENE). ( C ) Pecten maximus and the Hawaiian bobtail squid Euprymna scolopes : There
are multiple chromosomes in the squid genome that contain genes from ALG A1a, reflecting the
complex rearrangements of BCnS ALGs in the evolutionary history of cephalopods since the
common ancestor of all mollusks. ( Together
these networks show how the graph captures both orthology and genomic context, enabling
qualitative assessment of one-to-one versus split linkage relationships across species.

130

Darrin T. Schultz @dts.bsky.social · 19/08/2026

3. Two things you can click around to explore the genome data: I built an interactive version of the genome map. Hover any of the dots, search by taxon, lasso a region, export the table. This is useful for seeing how similar one genome is to others! atlas.evogeno.me

This shows a genome 2-dimensional scatter plot. A UMAP embedding of genome similarity across species. There is a web interface to search for species.

it is available at https://atlas.evogeno.me/

140

Darrin T. Schultz @dts.bsky.social · 19/08/2026

2. egt, the analysis software Every method in the paper is a subcommand: ALG fusion and dispersal inference, the topology embeddings, clade-specific locus pairs, the rate-through-time analyses, GO enrichment. Point it at your own clade! check the repo for more info. github.com/conchoecia/egt

diagram showing a phylogenetic tree on the left, with chromosomes remaining unfused in species 1 and species 2, but fused in species 3 and species 4. The next panel shows clusters of closely associated genes in the clade of species one plus species two, and a different cluster in the fused-and-mixed chromosome in the other clade of species 3 and species 4. This information is included in a distance embedding used to show distance-conserved loci in a genome. This information can also be used to embed whole genomes and describe them by similarity.

110

Darrin T. Schultz @dts.bsky.social · 19/08/2026

Everything we built for this is open-source! 1. The genome database. All 5,821 chromosome-scale assemblies, deduplicated, embargo-checked, and mapped onto the ancestral 29. Built with chrombase, so you can rebuild or extend it yourself. github.com/conchoecia/c...

github.com

GitHub - conchoecia/chrombase: Tools to build a database of chromosome-scale genomes

Tools to build a database of chromosome-scale genomes - conchoecia/chrombase

130

Darrin T. Schultz @dts.bsky.social · 19/08/2026

I also want to write a bit more about the technical side of the paper. This is a tree-of-life-scale dataset, so I needed to write special tools to work with the complete set of animal genomes available to us during the writing and revision process! There are 4 main resources you'll want:

130

Darrin T. Schultz @dts.bsky.social · 19/08/2026

So, just like the choices we make close the past and open new opportunities, chromosomal changes do the same. Fusions, inversions, and other changes wreck local interactions, but build the new neighborhoods that new regulation can be built on.

130

Darrin T. Schultz @dts.bsky.social · 19/08/2026

One example. A translation initiation gene and a V-ATPase subunit sit head to head, under 5 kb apart, in 90.4% of the 2,113 vertebrate species we checked. In humans, a single transcription factor (TF) binds between them. Outside vertebrates they are on the same chromosome but don't share that TF!

A plot comparing the genomic distance between one pair of genes across two large animal groups. In deuterostome genomes the pair sits at a consistently narrow separation, appearing as a tight spike. In spiralian genomes the same pair varies widely, from closely linked to many megabases apart.

150

Darrin T. Schultz @dts.bsky.social · 19/08/2026

Genes thrown together by a fusion-with-mixing or regular inversion can then stay together for hundreds of millions of years. We call this entanglement (sorry, physicists): a new neighborhood becomes hard to separate, plausibly because the new neighbors begin sharing regulation (but often neutral).

A diagram of a thoroughly mixed chromosome with red and blue genes interleaved. Curved arrows connect regulatory elements to genes from the other original chromosome, forming a web of cross-connections that would be broken if the mixing were reversed.

1104

Darrin T. Schultz @dts.bsky.social · 19/08/2026

Do heavily rearranged genomes eventually converge on some shared scrambled state? No. We embedded every genome by its internal architecture. Flies, nematodes, leeches, comb jellies and octopus are all thoroughly reshuffled, and each ends up somewhere different.

Twelve small copies of the same genome scatter plot, each highlighting one animal group in color against all other genomes in gray. Deuterostomes, protostomes, arthropods, molluscs, cnidarians, sponges and others each occupy their own separate region of the plot rather than overlapping.

150

Darrin T. Schultz @dts.bsky.social · 19/08/2026

One of my favorite tidbits from the paper is that species with a lot of chromosomes tend to have very few chromosomal fusions or ALG losses (see lamprey - right). How high chromosome numbers and whether fusions are selected against is a fantastic (albeit very expensive) next question!

Same alt text as the previous image, except highlighting that lamprey has many chromosomes and is shifted to the right.

150

Darrin T. Schultz @dts.bsky.social · 19/08/2026

Lineages tend to leave the ancestral karyotype in two opposite directions. Consolidation: fuse inward. True oysters have fused their way down to 10 chromosomes, and all 29 ancestral units are still intact inside them. Dissociation: break or duplicate outward. Octopus, leeches, vertebrates.

$A two-panel heatmap plotting animal genomes by their chromosome count. The upper blue panel shows the fraction of ancestral chromosome pairs that are fused; the lower red panel shows ancestral homology lost through gene dispersal. Labeled groups including nematodes, beetles, spiralians, moths, flies, teleost fish and birds occupy distinct regions.$

160

Darrin T. Schultz @dts.bsky.social · 19/08/2026

Chromosome evolution is also not a steady clock! While this topic deserves its own unique study and requires further calibration, we found differing bursts, rates, and dips of chromosome fusion and fissions (splits) across different clades in the animal tree.

Alt text: A chart of chromosome fusion rate through geologic time, running from about 700 million years ago to the present. The rate is high in the Ediacaran, peaking near 0.25 fusions per million years, then drops sharply through the Paleozoic and stays low for the rest of the record. This is likely due to branch sampling bias, however.

130

Darrin T. Schultz @dts.bsky.social · 19/08/2026

If you zoom in to what is happening in single chromosomes, there is a big spectrum of how much mixing within chromosomes (inversions) has happened. Some of the fastest are covered in other papers, like clitellate annelids. In this paper we discuss cephalopods (fast) and other mollusks (slow).

130

Darrin T. Schultz @dts.bsky.social · 19/08/2026

There are exactly 406 ways to pick two of those 29 ancestral units and fuse them (29 choose 2). Every single one has happened somewhere in the animal tree! At the whole-chromosome scale, evolution has sampled the entire menu.

A dense band of 406 horizontal tracks, one for each possible pair of ancestral chromosome units, plotted across the animal phylogeny. Colored marks show where each pair is found together on the same chromosome. Every track carries marks somewhere in the tree. There is an animal tree with thousands of genomes showing where ancestral linkages have fused or mixed.

1100

Darrin T. Schultz @dts.bsky.social · 19/08/2026

DNA doesn't "sit still" though. Rarely, two chromosomes fuse. Internal inverted pieces gradually shuffle the two halves into each other, like shuffling two decks of cards. We call it fusion-with-mixing, and once you shuffle, you can't un-shuffle. Sometimes they don't shuffle after fusing.

A diagram of two chromosomes, one red and one blue, fusing end to end and then, over successive time steps going down the page, becoming progressively interleaved by inversions until red and blue segments are thoroughly intermixed. A wedge alongside marks increasing entropy through time.

160

Darrin T. Schultz @dts.bsky.social · 19/08/2026

(Almost) Every animal alive today descends from an ancestor whose genome was built from around 29 ancestral units of linked genes. More than 600 million years later, those 29 are still detectable across all animals, and we can use them to reconstruct the history of chromosomes over time.

Two dot plots side by side, each comparing one animal's chromosomes against the 29 ancestral animal linkage groups. In both, that animal's chromosomes run along the bottom and the 29 ancestral units, labeled A1a, A1b, A2, B1 and so on, run up the side. Each colored dot is a gene, and a tight single-colored square means one whole chromosome corresponds to one ancestral unit.

Left: the lancelet Branchiostoma floridae, a small elongated fish-like animal, shown in silhouette. Its chromosomes form a clean staircase of tight colored squares up the diagonal, with almost nothing scattered off it.

Right: the jellyfish Rhopilema esculentum, also in silhouette. Its chromosomes form the same clean staircase, in the same order, with the same one-to-one correspondence.

1222

Darrin T. Schultz @dts.bsky.social · 19/08/2026

Our new paper in Science Advances asks how animals' genomes have changed since their origin from a common ancestor over 600 million years ago, and what patterns in their DNA have arisen in the churn of speciation and extinction. www.science.org/doi/10.1126/...

Five-panel diagram explaining fusion-with-mixing in chromosome evolution. Red and blue represent two different ancestral chromosomes throughout.

A, Fusion-with-mixing: a red chromosome and a blue chromosome fuse into one half-red, half-blue bar. Down the page, successive rows show inversions, drawn as X-shaped crossings between rows, progressively interleaving the two colors until the bottom bar is finely striped. A widening gray wedge labeled Entropy runs alongside an arrow marked Time plus evolution, pointing down.

B, Independent regulation: two separate chromosomes, one red and one blue. In each, a regulatory element, drawn as an oval, acts by an arrow on a gene within its own chromosome only.

C, FWM and novel regulation: on a fused, mixed chromosome, a red gene and a blue gene now sit next to each other across an inversion breakpoint. Red and blue arrows show regulatory elements reaching across it to act on the other chromosome's gene.

D, Regulatory entanglement: an extensively mixed chromosome, finely striped. Below it, a row of alternating red and blue genes is linked by a dense web of crossing curved arrows, so many overlapping regulatory connections that unmixing the chromosome would break them.

E, Emergence of entangled loci: a five-species phylogeny. A red and a blue chromosome fuse at an ancestral node, and helical icons on later branches mark continued mixing. Colored boxes mark four new red-blue regulatory interactions arising after inversions. Species A keeps its two chromosomes separate. In species B through E, the two earliest interactions, cyan boxes 1 and 2, are present in every descendant lineage, while later ones, yellow 3 and magenta 4, appear in fewer.

315055

Darrin T. Schultz @dts.bsky.social · 19/08/2026

I'll try to post here more now that I have a lab of my own, though my track record isn't encouraging. Email always works: dts@lehigh.edu

000

Darrin T. Schultz @dts.bsky.social · 19/08/2026

This fall I'm teaching Marine Biology to undergraduate and graduate students! Next semester is likely a genomics lab or seminar. This really is a dream job, in a department and at a university that have been remarkably generous, supportive, and professional to me and all other brand-new PIs.

100

Darrin T. Schultz @dts.bsky.social · 19/08/2026

My lab space in Iacocca Hall is nearly finished! We will be set up for single-cell/organismal genomics, cell culture, mouse work, and marine invertebrate culture.

100

Darrin T. Schultz @dts.bsky.social · 19/08/2026

I'm looking for students and postdocs with burning questions about fundamental biology -- people who want to generate their own data at the bench or in the field, and use our lab's computational foundation to answer them.

110

Darrin T. Schultz @dts.bsky.social · 19/08/2026

I'm thrilled to share that I've started as an Assistant Professor in Biological Sciences and the Oceans Research Center at Lehigh University! I'm building a group in evolutionary genomics to study the origin of animals, and to use those concepts to better understand human disease. More: evogeno.me

Darrin Schultz, a man with curly dark hair and tortoiseshell glasses, smiling at the camera in a brown blazer over a blue patterned shirt with sunglasses hooked on the collar, standing in front of a stone wall.

25912

Darrin T. Schultz @dts.bsky.social · 06/03/2026

I'm actively developing a Python 3 rewrite that will be installable via PyPI and conda. The goal is to make Dip-C even easier to use!

110

Darrin T. Schultz @dts.bsky.social · 06/03/2026

One thing I'm doing since joining the Tan Lab @ Stanford is working on the computational stack of 3D genome research. I put Dip-C v1.0 out as a release! This is an archive of the original Python 2 code from the single-cell 3D genomes in Science, Nat Struct Mol Biol, and Cell (2018–2023). 🧬

120

Reposted by Darrin T. Schultz

Pawel Burkhardt @pawelburkhardt.bsky.social · 04/03/2026

Even more excited — our paper is featured on the cover of Science Advances! Huge thanks to @alexandrejan.bsky.social who took this incredible photo of a ctenophore (aka comb jelly) and the editors. Here’s the cover 👇 www.science.org/doi/10.1126/...

39224

Darrin T. Schultz @dts.bsky.social · 05/03/2026

Stay tuned for more updates in the coming weeks. 🙂

000

Darrin T. Schultz @dts.bsky.social · 05/03/2026

I feel a lot of gratitude for Oleg Simakov, my labmates, and the Dept. of Neuroscience and Developmental Biology at @univie.ac.at the University of Vienna. We worked on understanding chromosome evolution across the history of animal evolution, and I grew a lot in that time.

130

Darrin T. Schultz @dts.bsky.social · 05/03/2026

A little late, but there is a fun new chapter in my life! I joined Longzhi Tan's lab at Stanford Neurobiology in November, and am now working on single-cell 3D architecture, neurodegeneration, and aging. Excited to bring a comparative genomics perspective to these topics. 🧬 Tan lab: 3dgeno.me

3dgeno.me

Tan Lab @ Stanford

Longzhi Tan's Laboratory of 3D Genomics at Stanford Neurobiology studies single-cell 3D genome architecture in development and aging by developing next-generation in vivo multi-omic assays and algorit...

140

Darrin T. Schultz @dts.bsky.social · 05/03/2026

When you zoom out from individual genes to consider the context of whole chromosomes across animal diversity, animal evolution takes on a different look. Oleg Simakov and I published a review in @annualreviews.bsky.social on topological approaches in comparative genomics. 🧬 doi.org/10.1146/annu...

074

Reposted by Darrin T. Schultz

Thibaut Brunet @thibautbrunet.bsky.social · 28/02/2026

Final version @nature.com of our paper describing unconventional multicellular development in a choanoflagellate inhabiting an extreme environment. A ton of new data since the first @biorxivpreprint.bsky.social preprint (which we've kept updating). A brief 🧵 (carried over from the old place)

16382152

Darrin T. Schultz @dts.bsky.social · 07/12/2023

Our paper on the octopus genome made the cover of G3! 🐙 Congrats to my mentee Dalila Destanović, Eve Seuntjens, and the CNAG team in Barcelona! Photo courtesy of David Stohlmann - a graduate student at the University of Vienna. academic.oup.com/g3journal/ar...

040