Search

Understanding how evolutionary pressures related to climate change have shaped the current genetic background of domestic animals is a fundamental pursuit of biology. Here, we generated whole-genome sequencing data from native goat populations in Iraq and Pakistan. Combined with previously published data on modern, ancient (Late Neolithic to Medieval periods), and wild Capra species worldwide, we explored the genetic population structure, ancestry components, and signatures of natural positive selection in native goat populations in Southwest Asia (SWA). Results revealed that the genetic structure of SWA goats was deeply influenced by gene flow from the eastern Mediterranean during the Chalcolithic period, which may reflect adaptation to gradual warming and aridity in the region. Furthermore, comparative genomic analysis revealed adaptive introgression of the KITLG locus from the Nubian ibex (C. nubiana) into African and SWA goats. The frequency of the selected allele at this locus was significantly higher among goat populations located near northeastern Africa. These results provide new insights into the genetic composition and history of goat populations in the SWA region.

Additional file 4

Additional file 2: Figure S1.Venndiagrams summarize unique and common variants among groups. Fig. S2. Theheat map of ChromoPainter���s coancestry matrix. Each row corresponds to therecipient genomes and columns represent the donor individuals. Fig. S3. Cross validationerror (CV) plot from ADMIXTURE. Fig. S4. Box plots of nucleotide diversity, calculated in 50kb sliding window with 20 kb increments across the genome. Fig. S5. Genomic regions with selection signals in nativechickens by �� method. (A) Indigenousgroup versus Arian chickens. (B) Indigenous group versus White Leghornchickens. Fig. S6. Haplotype network based on pairwise differenceswithin the selective sweep region (Chr1: 176.34-176.36) inHSPH1.

Additional file 2: Figure S1. 17-mer count distribution for the goat genome size estimation. The 17-mer count distribution was used to estimate genome size. Note that the peaks around the depths of 36, 70 and 138 represent the heterozygous, homozygous and repeated Kmers, respectively. Figure S2. Comparison of assembly quality among various reference genome assemblies. The gap number and contig N50 of seven species were compared. The gap number and contig N50 (Mb) for each assembly are shown in the brackets. Figure S3. Read depth across chromosomes (top right panel) and unplaced scaffolds (main panel). The read depth of chromosomes and unplaced scaffolds was compared between Saanen_v1 and ARS1. The whole genome sequencing data of a Yunnan black goat (~ 40×) are mapped to Saanen_v1 and ARS1. The read depth was calculated in 1-kb non-overlapping window. Figure S4. Venn diagram showing the intersection of identified genes among the 4104 single-copy orthologs in mammalia_obd9 database for BUSCO analysis. The intersection of identified genes from BUSCO analysis is shown for the four genome assemblies. Figure S5. Whole-genome alignment between Saanen_v1 and ARS1. The collinearity between Saanen_v1 and ARS1 is shown by whole-genome alignment. The Y chromosome and scaffolds of the two assemblies were excluded from the alignments. Figure S6. Structural variations detected in Saanen_v1 as compared with ARS1. The figure was generated by Assemblytics, displaying the summary statistics of structural variations. Figure S7. Hi-C contact matrix of ARS1 supports that the discrepancy between the alignments is likely due to assembly errors in ARS1. The assembly errors in ARS1 was evidenced by the Hi-C contact matrix. For each putative error region, the Hi-C heatmaps from ARS1 (left panel) and Saanen_v1 (right panel) were shown with the arrows indicating the discordant signals potentially caused by incorrect assembly. Figure S8. Alignment of the Saanen_v1 assembly and the sheep genome for the regions surrounding chr11:103,733,339–106,912,333 bp. The agreement between Saanen_v1 and sheep suggested that the corresponding region in ARS1 was incorrect. Figure S9. Sequence alignment between Saanen_v1 and CHIR_2.0 for the eight putative regions with assembly errors. Most of the assembly errors could be confirmed by collinearity between Saanen_v1 and CHIR_2.0 in the corresponding regions. Figure S10. The Hi-C heatmap shows that the putative Y and X chromosomes reside on the proximal and distal ends of PAR. We used Hi-C heatmap to infer X and Y chromosomes by locating their shared PAR region.

Additional file 1: Table S1. Candidate genes related to production traits identified through signature selection analysis in sheep.

Additional file 1. Supplementary methods for Hi-C library preparation and verification of previously reported assembly errors in ARS1

Additional file 3: Table S1. The publicly available miRNA-seq data used for miRNA annotation. We downloaded publicly available miRNA-seq data to annotate miRNA genes. Table S2. Summary of raw reads from PacBio sequencing. The data summarized the reads counts and length of PacBio sequencing. Table S3. Comparison of the basic statistics of the Saanen_v1 and ARS1 assemblies. The assembly length and continuity were compared between Saanen_v1 and ARS1. Table S4. Comparison of the repeat content of Saanen_v1 with ARS1. The total repeat content of Saanen_v1 was slightly higher than ARS1. The unplaced scaffolds from each assembly were not included for comparison. Table S5. Telomere signals identified in each assembly. Telomeres were found on 27 chromosomes of Saanen_v1 compared with 7, 6 and 5 chromosomes of ARS1, sheep (Oar_rambouillet_v1.0) and cattle (ARS-UCD1.2), respectively. Table S6. Centromere signals identified in each assembly. Saanen_v1 harbors more and longer centromeric repeats per chromosome than the other three assemblies of the reference genome for goat (ARS1), sheep (Oar_rambouillet_v1.0) and cattle (ARS-UCD1.2). Table S7. Structural inconsistencies when comparing Saanen_v1 and ARS1 assemblies. Various categories of structural inconsistencies were compared between Saanen_v1 and ARS1 using the FRC_align tool. Table S8. Mapping ratio of whole-genome sequencing data. Four whole-genome sequencing data of domestic goats were aligned to Saanen_v1 and ARS1 to compare the mapping ratio. Table S9. Mapping ratio of RNA-seq data. Nine RNA-seq datasets were aligned to Saanen_v1 and ARS1 to compare the mapping ratio. Table S10. Mapping ratio of Iso-seq data. Four Iso-seq datasets were aligned to Saanen_v1 and ARS1 to compare the mapping ratio. Table S11. Summary of structural variations in Saanen_v1 as compared with ARS1. The data represented the count and total length of structural variations in Saaenen_v1 as compared with ARS1. Table S13. Blast alignment of previously reported goat Y chromosome amplicons to the Y chromosome assembly included in Saanen_v1. The previously reported goat Y chromosome amplicons could be aligned to our generated Y chromosome assembly, implying the validity of our sequence. Table S15. Information on the large gap regions (> 1 Mb) between two adjacent SNP probes. The data presented seven large gap regions which were not covered by SNP probes.