We used the gadmor2 assembly (49) as a reference. We aligned the fastq reads to the reference using bwa mem (51) and used samtools (52) to generate sorted and indexed bam files from the sam files. We used picard (http://broadinstitute.github.io/picard) to merge bam files from different lanes, mark and remove duplicates, and build index after using Genome Analysis Toolkit (GATK) for indel realignment in accordance with GATK best practices (53). We separately aligned the fastq files to the Atlantic cod mtDNA genome using the same methods.

To determine the tree topology of the taxa based on the nuclear genome, we used ngsDist (54) to estimate pairwise genetic distances among taxa for each linkage group. The average of the linkage group distances weighted by the length of the linkage group in nucleotides yielded the whole-genome distance. A neighbor-joining tree was made using ape and package in R (55). To determine the tree topology based on mtDNA, we generated sequences in fasta format using ANGSD doFasta 3, which calls bases by the effective base depth. GenBank sequences for Polar cod whole-genome mtDNA from GenBank accession no. AM919429 and Blue whiting (M. poutassou) whole-genome mtDNA from GenBank accession no. FR751401 were added and aligned using muscle. Neighbor-joining trees were made with ape.

For formal tests of admixture and introgression, we used ANGSD (31) (version 0.921-8-gc12d2fa) to do the ABBA-BABA D statistic test of ancient admixture (33) using the multipopulation version (−doAbbababa2 1), allowing multiple individuals for each group (56). The ABBA-BABA test is based on a quartet of taxa H1, H2, H3, and H4 arranged in a tree (((H1,H2)H3) H4), with H4 as the outgroup. Here, the outer taxon H3 is a potential introgressor, and the inner taxa H1 and H2 are potential recipients of introgression from H3. If the outgroup is set to state A and the outer taxon (H3) to derived state B, then the inner taxa (H1,H2) are expected to show the AB and BA pattern equally frequently under incomplete lineage sorting. The genome was scanned, and the number of ABBA and BABA patterns were counted. The test statistic D is the standardized difference D = nABBA − nBABA, with an expectation of D = 0 under the null hypothesis of lineage sorting. A significant deviation from the expectation of D = 0 was then taken as evidence for admixture. A negative value of D means that the H1 potential recipient is closer to the H3 potential introgressor than H2 is to H3. A positive value of D means that H2 is closer to H3 than H1 is to H3. Statistical significance of the normalized test statistic Z score was estimated with a block jackknife (56). To account for multiple testing, we adjusted P values using the method of (57). We used the Arctic cod as the outgroup (Fig. 2B) and combined the Atlantic cod data into one group. The ABBA-BABA test was done for each linkage group and for the whole genome. On the basis of the mtDNA genome (Fig. 2A), the Polar cod specimen clustered with a Blue whiting mtDNA genome sequence as an outgroup to our taxa. Therefore, we also ran the ABBA-BABA test with Polar cod as the outgroup (table S2). Transitions are common in DNA data, and they may be homoplasies due to repeat mutations. All ABBA-BABA results presented here were transversions with transitions removed. The number of transversion sites is given in tables S1, S2, and S3. Plots of the fit of the ABBA-BABA D to the classical phylogeny were made using the admixturegraph package (58).

We also used the dfoil (github.com/jbpease/dfoil) software to detect introgression in a five-taxon symmetric phylogeny (59). The DFOIL (first, outer, inner, and last) are a set of ABBA-BABA–like D statistics that can facilitate detection of the polarization of introgression among taxa. A diagnostic pattern with at least one nonsignificant D indicates the direction of introgression (59).

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.