3.1 The majority of the two-repeat units of CG-rich trinucleotides resided in colonies
In total, 609,792 two-repeat units were detected genome-wide, of which 462,792 (75.89%) were arranged in 81,118 colonies (p <0.05) (Fig.1).
These colonies were spread across all chromosomes (Fig. 2a). Chromosome 1 showed a higher colony count (n = 6887) compared to other chromosomes. To find the percentage of colony distribution across chromosomes, the number of colonies in each chromosome was normalized with the length of that chromosome. Our results indicated that chromosome 1 had the highest percentage of colonies, followed by chromosomes 19, 17, and 22 (Fig.2b).
Several of the colonies that were further studied, were extensively compound, and coincided with colonies of other compound formulas in primates, and in a number of instances even in mouse (Table 1). Across the colonies, the two-repeat units were either pure or overlaps of two or more pure units.
3.2. Several of the large and medium-size colonies coincided, with extensive dynamicity, in primates
Several of the large and medium-sized colonies were also detected in other primates (Table 1).
Table 1. Several large and medium-sized colonies in human and their corresponding colonies in primates and mouse.
Colony Formulab
|
Chr. No.
|
Colony interval
|
Location
|
Gene
|
Colony Sizea
|
Human
|
Chimpanzee
|
Gorilla
|
Macaque
|
Mouse Lemur
|
Mouse
|
Transcript ID
|
C266
|
[(GGC)2]148
[(GCG)2]118
|
[(GGC)2]50
[(GCG)2]37 [(CGG)2]4
|
](GGC)2[142
](GCG2)[138
](CGG)2[2
|
|
|
|
22
|
22:20354460-20361885
|
Intron 7 of a novel gene
|
ENSG00000290950.1
|
ENST00000486536.6
|
C223
|
[(GGC)2]28
[(GCC)2]33
[(GCG)2]30
[(CGC)2]45
[(CCG)2]49
[(CGG)2]38
|
[(GGC)2]3 [(GCC)2]10 [(GCG)2]2 ](CGC)2[9 ](CCG)2[14 ](CGG)2[5
|
[(GGC)2]3 [(GCC)2]6
[(GCG)2]1 ](CGC)2[5
](CCG)2[9
](CGG)2[3
|
[(GGC)2]2
[(GCC)2]10 [(GCG)2]3 ](CGC)2[5 ](CCG)2[5 ](CGG)2[3
|
|
[(GGC)2]4
[(GCC)2]1 [(GCG)2]2 ](CGC)2[2 ](CCG)2[2 ](CGG)2[3
|
21
|
21:8249244 -8259593
|
19 kb downstream of a novel gene
|
ENSG00000278996.1
|
ENST00000623664.1
|
C210
|
[(GGC)2]52
[(GCG)2]84 [(CGC)2]5
[(CGG)2]69
|
[(GGC)2]5
[(GCG)2]7
](CGG)2[6
|
[(GGC)2]3
[(GCC)2]2
[(GCG)2]7
[(CGG)2]7
|
|
|
|
2
|
2:90323162 -90326439
|
640 kb downstream of a novel gene
|
ENSG00000283659.1
|
ENST00000636451.1
|
C134
|
[(GGC)2]29
[(GCC)2]21
[(GCG)2]26
[(CGC)2]18
[(CCG)2]17
[(CGG)2]23
|
[(GGC)2]5
[(GCC)2]8
[(GCG)2]7
[(CGC)2]4
[(CCG)2]4
[(CGG)2]6
|
|
|
[(GGC)2]4
[(GCC)2]6
[(GCG)2]3
[(CGC)2]8 [(CCG)2]6
[(CGG)2]2
|
[(GGC)2]3
[(GCC)2]2
[(GCG)2]2
[(CGC)2]4
[(CGG)2]2
|
13
|
13:109782140 -109787586
|
Promoter, 5 UTR, EXON 1 of the IRS2 gene
|
ENSG00000185950.9
|
ENST00000375856.5
|
C133
|
[(GGC)2]42
[(GCG)2]51 [(CGG)2]40
|
[(GCC)2]1 [(GCG)2]1 ](CGC)2[3 ](CCG)2[2
|
|
|
|
|
8
|
8:675038 -676522
|
Intron 1 of the ERICH1 gene
|
ENSG00000104714.14
|
ENST00000522706.5
|
C130
|
[(GGC)2]67
[(CGC)2]8
[(CCG)2]14
[(CGG)2]41
|
](GCC)2[9 ](CCG)2[6
](CGG)2[1
|
](GCC)2[6 ](CCG)2[5
](CGG)2[1
|
](GCC)2[3
](CCG)2[1
|
|
|
7
|
7:63748479 -63765529
|
1.5kb downstream of the CICP24 gene
|
ENSG00000271466.1
|
ENST00000605464.1
|
C123
|
[(GGC)2]17
[(GCC)2]13 [(GCG)2]20 [(CGC)2]24 [(CCG)2]24 [(CGG)2]25
|
[(GGC)2]18
[(GCG)2]26
](CGG)2[24
|
[(GGC)2]4
[(GCC)2]1
[(GCG)2]9
](CGC)2[1
](CCG)[1
](CGG)2[7
|
|
[(GGC)2]4
[(GCC)2]2
[(GCG)2]4
[(CGC)2]4 [(CCG)2]2 [(CGG)2]4
|
|
21
|
21:8438626 -8442627
|
147 KB downstream of a novel gene
|
ENSG00000280441.3
|
ENST00000651813.1
|
C119
|
[(GGC)2]17 [(GCC)2]11 [(GCG)2]21 [(CGC)2]24 [(CCG)2]22 [(CGG)2]24
|
](GGC)2]1 [(GCC)2]1 ](CGC)2[1 ](CCG)2[3 ](CGG)2[1
|
|
](GGC)2]4 [(GCC)2]2 [(GCG)2]3 ](CGC)2[4 ](CCG)2[5 ](CGG)2[4
|
](GGC)2]2 [(GCC)2]2 [(GCG)2]1 ](CGC)2[2 ](CCG)2[2 ](CGG)2[3
|
|
21
|
21:8211381 -8215364
|
Intron 2 of a novel
Gene
|
ENSG00000278996.1
|
ENST00000623664.1
|
C119
|
[(GGC)2]16 [(GCC)2]12 [(GCG)2]20 [(CGC)2]24 [(CCG)2]24 [(CGG)2]23
|
](GGC)2]4 [(GCC)2]2 [(GCG)2]3 ](CGC)2[3 ](CCG)2[5 ](CGG)2[2
|
[(GGC)2]2 [(GCC)2]1 [(GCG)2]1 ](CGC)2[1 ](CCG)[2
](CGG)2[1
|
](GGC)2]4 [(GCC)2]2 [(GCG)2]3 ](CGC)2[5 ](CCG)2[5 ](CGG)2[4
|
|
|
21
|
21:8394416-8398398
|
137 kb downstream of a novel gene
|
ENSG00000280441.3
|
ENST00000623860.3
|
C110
|
[(GGC)2]52
[(CGG)2]58
|
|
|
](GCC)2[5 ](CCG)2[4
|
[(GCC)2]3
[(GCG)2]1
[(CGC)2]2 [(CCG)2]2 [(CGG)2]3
|
|
6
|
6:167421795 -167424368
|
37 kb upstream of the TCP10L3 gene
|
ENSG00000290751.1
|
ENST00000366827.6
|
C96
|
[(GGC)2]12 [(GCC)2]23
[(GCG)2]8 [(CGC)2]21
[(CCG)2]25 [(CGG)2]7
|
](GGC)2]12
[(GCC)2]23
[(GCG)2]8
[(CGC)2]20
[(CCG)2]25
[(CGG)2]7
|
|
[(GGC)2]10 [(GCC)2]18 [(GCG)2]7 [(CGC)2]19 [(CCG)2]18 [(CGG)2]5
|
[(GGC)2]12
[(GCC)2]25
[(GCG)2]6 [(CGC)2]29
[(CCG)2]28
[(CGG)2]5
|
[(GGC)2]3
[(GCC)2]11
[(GCG)2]4 [(CGC)2]23
[(CCG)2]14 [(CGG)2]2
|
20
|
20:62064415-62066226
|
Promoter/Exon1 of the TAF4 gene
|
ENSG00000130699.20
|
ENST00000252996.9
|
C84
|
[(GGC)2]19 [(GCC)2]16 [(GCG)2]10 [(CGC)2]9 [(CCG)2]12 [(CGG)2]9
|
[(GGC)2]18 [(GCC)2]14 [(GCG)2]10 [(CGC)2]9 [(CCG)2]10 [(CGG)2]9
|
[(GGC)2]8
[(GCC)2]1
[(GCG)2]4
[(CCG)2]1 [(CGG)2]3
|
[(GGC)2]17 [(GCC)2]14
[(GCG)2]8 [(CGC)2]10 [(CCG)2]10 [(CGG)2]7
|
[(GGC)2]5 [(GCC)2]8 [(GCG)2]1 [(CGC)2]5 [(CCG)2]4 [(CGG)2]2
|
[(GGC)2]3 [(GCC)2]8 [(CGC)2]1 [(CCG)2]4 [(CGG)2]1
|
17
|
17:2055050 -2058974
|
Intron 1-Exon 2 of
the HIC1
|
ENSG00000177374.13
|
ENST00000619757.5
|
C77
|
[(GCG)2]25
[(CGG)2]52
|
](CCG)2[19
|
](CCG)2[7
|
|
|
|
4
|
4:189307407-189309555
|
16 kb upstream of a
novel gene
|
ENSG00000249877.1
|
ENST00000504057.1
|
C66
|
[(GGC)2]54 [(GCC)2]1 [(GCG)2]5 [(CGC)2]2 [(CGG)2]4
|
[(GGC)2]5 [(GCC)2]9 [(GCG)2]7 [(CGC)2]4 [(CCG)2]4 [(CGG)2]5
|
[(GCC)2]5
[(GCG)2]5
[(CGC)2]3
[(CCG)2]5
[(CGG)2]5
|
[(GGC)2]4 [(GCC)2]8 [(GCG)2]5 [(CGC)2]6 [(CCG)2]3 [(CGG)2]3
|
[(GGC)2]4
[(GCC)2]6
[(GCG)2]3
[(CGC)2]8 [(CCG)2]6
[(CGG)2]2
|
|
19
|
19:29898812-29902043
|
14 kb upstream of a
novel gene
|
ENSG00000278895.1
|
ENST00000624591.1
|
C62
|
[(CCG)2]62
|
](GCC)2[2 ](CGC)2[3 ](CCG)2[7
|
](CCG)2[3
|
](GCC)2[1 ](CCG)2[1
|
](GCC)2[1 ](CCG)2[1
|
|
8
|
8:1641877 -1644115
|
Intron 8 of the
DLGAP2 gene
|
ENSG00000198010.13
|
ENST00000637795.2
|
C56
|
[(GGC)2]1
[(GCG)2]1 [(CGG)2]54
|
|
|
|
|
|
7
|
7:158425196 -158428347
|
Intron 2 of the
PTPRN2 gene
|
ENSG00000155093.19
|
ENST00000389418.9
|
C55
|
[(GGC)2]1
[(GCC)2]47
[(GCG)2]1 [(CGC)2]3
[(CCG)2]3
|
](GGC)2]3 [(GCC)2]2 [(GCG)2]3 ](CGC)2[3 ](CCG)2[2 ](CGG)2[1
|
[(GGC)2]2
[(GCC)2]2
[(GCG)2]2 [(CGC)2]2 [(CCG)2]1 [(CGG)2]1
|
[(GGC)2]4 [(GCC)2]2 [(GCG)2]4 [(CGC)2]3 [(CCG)2]2 [(CGG)2]2
|
|
|
19
|
19:22925342 -22928128
|
66 kb downstream of
the ZNF723
gene
|
ENSG00000268696.3
|
ENST00000600766.3
|
aColony size, chromosomal location, colony interval, and gene and transcript IDs are based on the human genome, as reference. In instances where colonies were intergenic, the nearest gene to those colonies was annotated. The corresponding colonies in other primates and mouse were identified, using BLASTN. Instances in which the colonies were partially or not sequenced at the corresponding loci in a species, or were not detected in a particular species, were left blank.
bFormulas represent absolute numbers of units, regardless of being pure or overlapping. The table represents only a sample of the identified colonies. Based on Poisson distribution, the probability of the occurrence of the colonies in this table was inherent zero.
The largest colony in human was a compound colony of 266 units (C266), located on chromosome 22. This colony was detected of various formulas across human, chimpanzee, and gorilla (Fig. 3). The absolute number of the units, the distribution of the units in the pure and overlapping compartments, and thence the composition of the colonies was extensively dynamic across those species, adding multiple layers of complexity of the events across the colonies.
One type of pure unit, GGCGGC, was detected in human and chimpanzee, whereas two types of pure units, GGCGGC and GCGGCG, were detected in gorilla. The main overlapping unit across the colonies in human, chimpanzee, and gorilla was GGCGGCG, which resulted from unequal crossover between the two pure units, GGCGGC and GCGGCG, across evolution (please see section 3.4).
3.3. Conservation of several colonies beyond primates
Several of the colonies were not only conserved across primates, but also in mouse (Table 1), for example, C96 (Fig. 4). In comparison to C266, which was restricted to only two types of pure units, a composition of six pure units, and consequently, a more versatile repertoire of overlapping units were detected in C96 across primates and mouse. This is predictable, as this hotspot colony is more ancient than C266. While C96 was relatively conserved between human and chimpanzee, the colony was highly dynamic across other primates and mouse. We detected overlapping units that were specific to each species. For example, the complex overlapping unit, CGCCGCCCGGCGGCG, was human-specific, and a result of several unequal crossovers (please see section 3.4).
3.4. Models proposed for the recombination events in example colonies, C266, and C96
In C266 (Fig. 5A), the main overlapping unit, GGCGGCG, was the consequence of unequal crossover between two pure units, GGCGGC and GCGGCG. In a reverse process, i.e., unequal crossover between pure and overlapping blocks led to the gradual elimination of the pure unit, GCGGCG (yellow blocks) during the evolution of human and chimpanzee. Additional overlapping units were also detected in this hotspot colony in gorilla and chimpanzee, which were not detected in human, such as GGCGGCGG, which was the probable consequence of unequal crossover between three pure units of GGCGGC, GCGGCG, and CGGCGG, or the overlapping unit CGGCGGCG, which was the consequence of unequal crossover between CGGCGG, GGCGGC, and GCGGCG. It is conceivable that the pure unit, CGGCGG, emerged and extinguished during the evolution of C266 and its orthologous colonies in great apes.
In C96, more versatile pure units were detected, which led to more versatile overlapping units (5B and C). Some of the units incremented in density in primates, such as CCGCCGCC (Fig. 5B), or were species-specific, for example, the overlapping unit, CGCCGCCCGGCGGCG, which was detected in human only (Fig. 5C).
Unequal crossover at the two-repeat units is a mechanism for the emergence of CG-rich STRs
In some colonies, for example, in C96, consecutive unequal crossovers led to the emergence of STRs i.e., repeats of ≥3, such as (CCG)5 (Fig. 4 and Fig. 6A). That STR emerged in primates, and was not detected in mouse. In another remarkable example, (GCC)n emerged as a result of several consecutive recombination events, intermediate overlapping units of which were detectable across C96 and its orthologous colonies (Fig. 6B and C).