Takifugu rubripes

Whole genomic sequences of Danio rerio and Takifugu rubripes were retrieved from the ENSEMBL database [54 ]. Exon sequences with length > 800 bp were then extracted from the genome databases. The exons extracted were compared in two steps: (1) within-genome sequence comparisons and (2) between genome comparisons. The first step is designed to generate a set of single-copy nuclear gene exons (length > 800 bp) within each genome, whereas the second step should identify single-copy, putatively orthologous exons between D. rerio and T. rubripes (Figure 2). The BLAST algorithm was used for sequence similarity comparison. In addition to the parameters available in the BLAST program, we applied another parameter, coverage (C), to identify global sequence similarity between exons. The coverage was defined as the ratio of total length of locally aligned sequences over the length of query sequence. The similarity (S) was set to S < 50% for within-genome comparison, which means that only genes that have no counterpart more than 50% similar to themselves were kept. The similarity was set to S × > 70% and the coverage was set to C > 30% in cross-genome comparison, which selected genes that are 70% similar and 30% aligned between D. rerio and T. rubripes. Subsequent comparisons were performed on the newly available genome of stickleback (Gasterosteus aculeatus) and Japanese rice fish (Oryzias latipes), as described above. We programmed this procedure using PERL programming language to automate the processes and made the source code publicly available on our website [43 ]. We are in progress to make it available for other genomic sequences and parameter values.

Repetitive elements were identified de novo using RepeatModeler v2.0.1 (Flynn et al. 2020 (link)) with the “LTRStruct” option. RepeatMasker v4.1.1 (Tempel 2012 (link)) was used to screen known repetitive elements with two inputs: (1) the RepeatModeler output and (2) the vertebrata library of Dfam v3.3 (Storer et al. 2021 (link)). The resulting output files were validated and merged before redundancy was removed using GenomeTools v1.6.1 (Gremme et al. 2013 (link)). To identify and annotate candidate gene models, BRAKER v2.1.6 (Brůna et al. 2021 (link)) was used with mRNA and protein evidence. For annotation with BRAKER, the chromosome sequences were soft masked using the maskfasta function of BEDTools v2.30.0 (Quinlan 2014 (link)) with the “soft” option. Protein evidence consisted of protein records from UniProtKB/Swiss-Prot (UniProt Consortium 2021 (link)) as of 2021 January 11 (563,972 sequences) as well as selected fish proteomes from the NCBI database (A. ocellaris: 48,668, Danio rerio: 88,631, Acanthochromis polyacanthus: 36,648, Oreochromis niloticus: 63,760, Oryzias latipes: 47,623, Poecilia reticulata: 45,692, Stegastes partitus: 31,760, Takifugu rubripes: 49,529, and Salmo salar: 112,302). Transcriptomic reads from 13 tissues were used as mRNA evidence. These Illumina short reads were trimmed with Trimmomatic v0.39 (Bolger et al. 2014 (link)) as described above and mapped to the chromosome sequences with HISAT2 v2.2.1 (Kim et al. 2019 (link)). The resulting SAM files were converted to BAM format with SAMtools v1.10 (Li et al. 2009 (link)) and used as input for BRAKER. Of the resulting gene models, only those with supporting evidence (mRNA or protein hints) or with homology to the Swiss-Prot protein database (UniProt Consortium 2021 (link)) or Pfam domains (Mistry et al. 2021 (link)) were selected as final gene models. Homology to Swiss-Prot protein database and Pfam domains was identified using Diamond v2.0.9 (Buchfink et al. 2015 (link)) or InterProScan v5.48.83.0 (Zdobnov and Apweiler 2001 (link)), respectively. Functional annotation of the final gene models was completed using NCBI BLAST v2.10.0 (Altschul et al. 1990 (link)) with the NCBI non-redundant (nr) protein database. Gene Ontology (GO) terms were assigned to A. clarkii genes using the BLAST output and the “gene2go” and “gene2accession” files from the NCBI ftp site (https://ftp.ncbi.nlm.nih.gov/gene/DATA/). Completeness of the gene annotation was assessed with BUSCO v4.1.4 (actinopterygii_odb10) (Simão et al. 2015 (link)).

Tiger puffer juveniles (average initial body weight, 12.3 ± 0.5 g; average body length, 6.5 ± 0.5 cm) were purchased from Hongqi Modern Fishery Industrial Park (Rizhao, China), and transported to the Yellow Sea Aquaculture Co., Ltd. (Yantai, China), where the feeding trial was conducted. At the beginning of the experiment, 600 randomly selected healthy fish were divided into 15 polyethylene tanks (0.7 × 0.7 × 0.4 m). Each tank was stocked with 40 fish. Each diet was randomly fed to triplicate tanks. Fish were hand-fed to apparent satiation three times daily (7:00, 12:00, and 18:00). The feeding trial lasted 12 weeks (Figure 4). Following the first 8-week growing-out period, during which the five experiment diets were normally fed to experimental fish, was a four-week FO-finishing period, during which fish in all five groups were fed the FO control diet.
Sampling was conducted at the end of both the growing-out period (week 8) and the FO-finishing period (week 12). Before sampling, fish were fasted for 24 h. In each sampling point, four fish were randomly selected from each tank, and the muscle and liver samples were collected. The samples were immediately frozen with liquid nitrogen and then stored at −76 °C before use. All sampling protocols, as well as all fish rearing practices, were reviewed and approved by the Animal Care and Use Committee of Yellow Sea Fisheries Research Institute.

To compare the divergence among animals and among Saccharomyces species and lineages, we annotated a common set of single-copy orthologous genes with BUSCO v5.1.3 ^{162 (link)} using the eukaryota_odb10 database. We first downloaded the genome assemblies for Homo sapiens GRCh38p13 (GCA_000001405.28), Pan troglodytes ClintPTRv2 (GCA_002880755.3), Macaca mulata AG07107 (GCA_003339765.3), Mus musculus C57BL6J (GCA_000001635.9), Takifugu rubripes fTakRub1 (GCA_901000725.2), and Gallus gallus domesticus bGalGal1 (GCF_016699485.2). Then, we ran BUSCO on those genomes and high-quality genomes for Saccharomyces strains with no detected admixture or introgressions (Supplementary Data 1). Each organism’s amino acid sequences for each protein were pulled together. Amino acid alignments were performed using MAFFT with similar settings as above. Individual protein alignments were read in R with seqinr v4.2 package ¹⁶³ . Amino acid identity (AAI) values (Fig. 3c) for each protein between lineages of the same Saccharomyces species, between Saccharomyces species, and between the chosen animals were calculated using dist.alignment function implemented in seqinr with the “matrix=identity” setting. Mean AAI values for each comparison was plotted with ggplot2.