A phased genome assembly of a Colombian Trypanosoma cruzi TcI strain and the evolution of gene families

Abstract

Chagas is an endemic disease in tropical regions of Latin America, caused by the parasite Trypanosoma cruzi. High intraspecies variability and genome complexity have been challenges to assemble high quality genomes needed for studies in evolution, population genomics, diagnosis and drug development. Here we present a chromosome-level phased assembly of a TcI T. cruzi strain (Dm25). While 29 chromosomes show a large collinearity with the assembly of the Brazil A4 strain, three chromosomes show both large heterozygosity and large divergence, compared to previous assemblies of TcI T. cruzi strains. Nucleotide and protein evolution statistics indicate that T. cruzi Marinkellei separated before the diversification of T. cruzi in the known DTUs. Interchromosomal paralogs of dispersed gene families and histones appeared before but at the same time have a more strict purifying selection, compared to other repeat families. Previously unreported large tandem arrays of protein kinases and histones were identified in this assembly. Over one million variants obtained from Illumina reads aligned to the primary assembly clearly separate the main DTUs. We expect that this new assembly will be a valuable resource for further studies on evolution and functional genomics of Trypanosomatids.

Figure 1

(A) Average read depth per contig of HiFi reads aligned to the primary haplotype and to the complete phased assembly. (B) Histogram of relative allele dosage in sites with two observed alleles after mapping to the haplotype H1 or to the combined assembly. (C) Violin plots of relative allele dosage in selected individual chromosomes. (D) Synteny-based alignment between the assembled haplotypes of the genome of Dm25 and between the haplotype H1 and the Brazil A4 genome assembly. (E) Zoom into the contigs assigned to chromosomes 30, 31, and 32 of Dm25.

Figure 2

Distribution of genes and repeat elements in the primary assembly of the Dm25 strain. (A) Karyotype representation. Contigs with suffixes c1 or c2 have telomere repeats on only one side. Grey bands represent the disruptive compartment. (B) GC-content over 10 kbp windows. Limits: 40–60%. (C) Average read depth in windows of 100 kbp. (D) Density of transposable elements. (E–G) Density of (E) gene families making the disruptive compartment (Transialidases, MASP and Mucin), (F) single copy genes. (G) Other known repeat families (DGF, GP63, RHS and TED). (H) Gene orientation, zero for negative and one for positive. Lines represent strand switches. (I) Density of protein kinases and (J) Density of histones. (K,L) Distributions of synonymous nucleotide divergence rate (ks) and relative non-synonymous to synonymous divergence rate (ka/ks) for pairs of orthologs, comparing the primary assembly of Dm25 (H1) with (K) other T. cruzi assemblies and (L) with assemblies of other species.

Figure 3

(A) Distributions of synonymous nucleotide divergence rate (ks) and relative non-synonymous to synonymous divergence rate (ka/ks) for pairs of paralogs of the six main multicopy gene families, and for protein kinases (PKS) and histones (HIS), discriminating tandem (close) paralogs from distant paralogs. (B) Alignment of the contig assigned to chromosome 32 of the haplotype H1 of Dm25 with the contig assigned to chromosome 32 of Brazil A4. The lower track highlights a tandem array of protein kinases spanning the black rectangle. The numbers besides the alignment indicate the number of copies annotated in the regions. (C) Alignment of the contig assigned to chromosome 4 of the haplotype H1 of Dm25 with the contig assigned to chromosome 4 of Brazil A4. The lower track highlights a tandem array of histones spanning the black rectangle. The numbers besides the alignment indicate the number of copies annotated in the regions.

Figure 4

Genomic variants identified between Illumina reads of T. cruzi strains from different DTUs and our haploid T. cruzi assembly (H1). (A) Number of homozygous and heterozygous variants in T. cruzi strains. (B) Neighbor joining clustering of genetic distances between T. cruzi strains from different DTUs, including the haploid Dm25 assembly. (C) close up of the variability within TcI. The names indicate the DTU—country of isolation—strain. Bra Brazil, Bol Bolivia, Pan Panama, Chi Chile, Col Colombia, Ven Venezuela, Ecu Ecuador.