1/1
9 files

Data from: Gene family innovation, conservation and loss on the animal stem lineage

dataset
posted on 28.05.2018, 19:45 authored by Daniel RichterDaniel Richter, Parinaz Fozouni, Michael EisenMichael Eisen, Nicole King
Choanoflagellates, the closest living relatives of animals, can provide unique insights into the changes in gene content that preceded the origin of animals. However, only two choanoflagellate genomes are currently available, providing poor coverage of their diversity. We sequenced transcriptomes of 19 additional choanoflagellate species to produce a comprehensive reconstruction of the gains and losses that shaped the ancestral animal gene repertoire. We identified ~1,944 gene families that originated on the animal stem lineage, of which only 39 are conserved across all animals in our study. In addition, ~372 gene families previously thought to be animal-specific, including Notch, Delta, and homologs of the animal Toll-like receptor genes, instead evolved prior to the animal-choanoflagellate divergence. Our findings contribute to an increasingly detailed portrait of the gene families that defined the biology of the Urmetazoan and that may underpin core features of extant animals.

Dataset 1. Final sets of contigs from choanoflagellate transcriptome assemblies. There is one FASTA file per sequenced choanoflagellate. We assembled contigs de novo with Trinity, followed by removal of cross-contamination that occurred within multiplexed Illumina sequencing lanes, removal of contigs encoding strictly redundant protein sequences, and elimination of noise contigs with extremely low (FPKM < 0.01) expression levels. [This Dataset is identical in version 1.]

Dataset 2. Final sets of proteins from choanoflagellate transcriptome assemblies. There is one FASTA file per sequenced choanoflagellate. We assembled contigs de novo with Trinity, followed by removal of cross-contamination that occurred within multiplexed Illumina sequencing lanes, removal of strictly redundant protein sequences, and elimination of proteins encoded on noise contigs with extremely low (FPKM < 0.01) expression levels. [This Dataset is identical in version 1.]

Dataset 3. Expression levels of assembled choanoflagellate contigs. Expression levels are shown in FPKM, as calculated by eXpress. Percentile expression rank is calculated separately for each choanoflagellate. [This Dataset is identical in version 1.]

Dataset 4. Protein sequences for all members of each gene family. This includes sequences from all species within the data set (i.e., it is not limited to the choanoflagellates we sequenced). [This Dataset is identical in version 1.]

Dataset 5. Gene families, group presences, and species probabilities. For each gene family, the protein members are listed. Subsequent columns contain inferred gene family presences in different groups of species, followed by probabilities of presence in individual species in the data set. [This Dataset is identical in version 1.]

Dataset 6. List of gene families present, gained and lost in last common ancestors of interest. A value of 1 indicates that the gene family was present, gained or lost; a value of 0 indicates that it was not. The six last common ancestors are: Ureukaryote, Uropisthokont, Urholozoan, Urchoanozoan, Urchoanoflagellate and Urmetazoan. Gains and losses are not shown for the Ureukaryote, as our data set only contained eukaryote species and was thus not appropriate to quantify changes occurring on the eukaryotic stem lineage. [This Dataset was updated from version 1.]

Dataset 7. Pfam, transmembrane, signal peptide, PANTHER and Gene Ontology annotations for all proteins. Annotations are listed for all proteins in the data set, including those not part of any gene family. Pfam domains are delimited by a tilde (~) and Gene Ontology terms by a semicolon (;). Transmembrane domains and signal peptides are indicated by the number present in the protein, followed by their coordinates in the protein sequence. [This Dataset is identical in version 1.]

Dataset 8. Pfam, transmembrane, signal peptide, PANTHER and Gene Ontology annotations aggregated by gene family. The proportion of proteins within the gene family that were assigned an annotation is followed by the name of the annotation. Multiple annotations are delimited by a semicolon (;). [This Dataset is identical in version 1.]

Dataset 9. Modifications to our sequences for upload to the NCBI Transcriptome Shotgun Assembly database. All contigs that were either modified or removed are listed. For contigs entirely excluded (“excluded”) or trimmed (“trimmed”) during the TSA submission process, the source of the identified adapter or bacterial sequence is provided. Coordinates for sequence trimming are given with respect to the unmodified contig sequence. In cases where a contig was trimmed (but not removed) and the trimming caused a change in its predicted protein sequence, the modified sequence is provided. “Short” or “trimmed;short” contigs were removed because they were below the minimum length of 200 bases (after trimming) imposed by the TSA. [This Dataset was added from version 1.]

Daniel J Richter, Parinaz Fozouni, Michael Eisen, Nicole King. Gene family innovation, conservation and loss on the animal stem lineage. eLife 2018;7:e34226

Funding

NIH R01 GM089977

History