We screened the LLOYNC03 "M" (Lawrence Livermore National Laboratory) Y-chromosomal library (5.5xY chromosome coverage) for CHEK2 duplicon containing cosmids using a Y-chromosomal probe generated by the following PCR-conditions: 95°C for 5 min, 40 cycles of 94°C for 30 sec, 62°C for 30 sec and 72°C for 1 min and finally 72°C for 5 min. The following primers were used to generate a 575 bp probe: IP-cos-SD1-for 5'-ACCCCCTTAGTAGCGTCCTTAGCTC-3' and IP-cos-SD1-rev 5'-ACCACCGGAGTTTCACAAAGAAAGT-3'. The gel-purified probe was radioactively labelled according to the 'random priming' protocol 35 . Prehybridization and hybridization of the high density gridded filters were carried out for 18 hrs at 42°C according to the manufacturer's protocol. Final filter-washing was carried out in 1%SDS/2xSSC solution for 1 hr at 65°C. Primers located proximal and distal of the derivative Y-chromosomal CHEK2 duplicon were used for PCR (conditions see above) to identify cosmids containing the entire duplicon:

PB-cos-SD1-for 5'-AGCGCAAATTGCAGAATTACAAAGA-3', PB-cos-SD1-rev 5'-GGTTAGAGAGGATAAGCCGCATGTT-3', DB-cos-SD1-for 5'-GATCCCGCACATTTGTTCATTAGAG-3', DB-cos-SD1-rev 5'-CAAAAGCTTGAATTCTGTGCCTCAGT-3'

We used RepeatMasker http://www.repeatmasker.org/ and Tandem Repeats Finder 36 to search for CAGGG repeat sequences within the Y-chromosomal CHEK2 duplicon containing cosmid LLOYNC03"M" 22E01. Both analyses failed to identify CAGGG repeat sequences in the derivative CHEK2 duplicon and in both adjacent derivative duplicons, which are the IGL@- (Immunoglobulin lambda@ locus-) and the NHEDC1- (Na⁺/H⁺ exchanger domain containing 1-) duplicons. Nevertheless, RepeatMasker and Tandem Repeats Finder analysis revealed the presence of extensive CAGGG repeat sequences within the ancestral IGL@ locus in 22q11.21 (NT_011520: 1994740–2060046). Subsequent pairwise alignment of both ancestral and the derivative Y-chromosomal IGL@ loci disclosed a more diverged CAGGG repeat like sequence. By using the DnaSP Ver.4.10.9 software 72% nucleotide sequence identity was detected between the CAGGG repeats present within the ancestral IGL@-duplicon (NT_011520: 2033670–2035327) and the derivative CAGGG repeats in Yq11.1/Yq11.21 (NT_113819: 398233–403886).

The following blood samples from non-related individuals were used for qPCR analysis: blood samples from the chimpanzees (Pan troglodytes, PTR) Max (LN: 444) and Fritz (LN: 776, wild-born) were obtained from the Zoologisch-Botanischer Garten Wilhelma Stuttgart (Germany). The blood sample from chimpanzee Marcel (LN: 39) was obtained from TNO Primate Centre Rijswijk (Netherlands). Lowland gorilla (Gorilla gorilla gorilla, GGO) blood samples of Jangu (LN: 315), Gaidi (LN: 916, wild-born) and Fritz (LN: 673, wild-born) were obtained from the Zoo Wuppertal, the Zoo Leipzig and the Tiergarten Nürnberg, respectively. Two blood samples of the Bornean orangutans (Pongo pygmaeus pygmaeus, PPY) Thai (LN: 866) and Napoleon (LN: 1005) were obtained from the Zoo Duisburg (Germany) and the Zoo Studen (Switzerland), respectively. Blood samples from two rhesus macaque individuals (Macaca mulatta, MMU; LN: 1053 and 1054) and one hamadryas baboon individual (Papio hamadryas, PHA; LN: 496) were obtained from the Deutsches Primaten Zentrum Göttingen (Germany).

Skin tissue of a pig-tailed macaque (Macaca nemestrina, MNE) was provided by the Deutsches Primaten Zentrum Göttingen (Germany) and was used to establish a fibroblast cell line. For each species one of the above listed blood samples and the fibroblast cell line were used for FISH analysis. Lymphoblastoid cell lines of the white-cheeked crested gibbon (Nomascus leucogenys, NLE) and the common marmoset (Callithrix jacchus, CJA) were kindly provided by S. Müller (Munich) and were used for FISH analysis.

FISH analysis of metaphase spreads derived from lymphocytes or lymphoblastoid and fibroblast cell lines from non-related human (Homo sapiens, HSA) and non-human primate males was performed. Prior to FISH, the slides were treated with RNase followed by pepsin digestion as described 37 . FISH was carried out following the protocol described previously 38 . Chromosome in situ suppression was applied to clones from the human fosmid library WI-2 (WI2-1621D20, WI2-819H21) and from the Y chromosome specific cosmid library LLOYNCO3"M" (LLOYNCO3"M"22E01). Human whole-chromosome painting (WCP) libraries 39 were used to unequivocally assign hybridizing signals to orthologous regions in lesser apes, Old World monkeys and New World monkeys. pMR100, a mouse-derived rDNA-containing plasmid, was used to tag the Old World monkey marker chromosome. After FISH the slides were counterstained with DAPI (4',6-diamidino-2-phenylindole; 0.14 μg/ml) and mounted in Vectashield (Vector Laboratories). Preparations were evaluated using a Zeiss Axiophot epifluorescence microscope equipped with single-bandpass filters for excitation of red, green, and blue (Chroma Technologies, Brattleboro, VT). During exposures, only excitation filters were changed allowing for pixel-shift-free image recording. Images of high magnification and resolution were obtained using a black-and-white CCD camera (Photometrics Kodak KAF 1400; Kodak, Tucson, AZ) connected to the Axiophot. Camera control and digital image acquisition involved the use of an Apple Macintosh Quadra 950 computer.

FASTA formatted sequence files used to generate phylogenetic trees were extracted from the corresponding GenBank accession numbers. Sequence alignments were built by using CLUSTALW (version 1.82) 40 , and neighbor-joining phylograms created by using MEGA (Molecular Evolutionary Genetic Analysis) v4.0 http://www.megasoftware.net 41 . Neighbor-joining analysis was used with complete deletion parameters and bootstrap (1,000 iterations) to provide confidence of each branching point in the phylogenetic trees. Neighbor-joining methods were chosen as they are amenable to calculating divergence times between sequence taxa. We estimated the number of substitutions per site per year by correcting the divergence times for multiple substitutions using Kimura's two-parameter model 42 . As the rates of nucleotide substitution vary for pseudogenic sequences, the rate of nucleotide substitution was calibrated based on orthologous sequence comparisons using a divergence of 25 Mya for macaque-human divergence 43 . Duplication timing events were calculated by applying the equation r = k/2 T 44 , where r is the rate of nucleotide changes per bp per yr, k is the distance calculated between the ancestral and paralogous sequences, and T is the time of divergence of the molecules.

Interspecies quantitative PCR was carried out using primers specific for CHEK2 exon 14. Primers were designed with the assistance of the Promega Plexor Primer Design Software. The following primer sequences were used: CHEK2-exon14-F 5'-GGACCTTGTCAAGAAGTTGTTGGT-3', CHEK2-exon14-R 5'-GGTGTCTTAAGGCTTCTTCTGTCGTA-3', CFTR-F 5'-CGCGATTTATCTAGGCATAGGC-3' and CFTR-R 5'-TGTGATGAAGGCCAAAAATGG-3'

We used the ABI Prism 7900 HT system (Applied Biosystems) for real time detection. Reactions contained 0.25 μM of each primer and 5 μl of QuantiTec SYBR^® Green PCR Master Mix (Quiagen) in a total of 10 μl. Assays included DNA standards at a final concentration of 5.0, 2.5, 1.25, 0.625 and 0.3125 ng/μl, a no-template control, or 1 ng/μl of the species DNA in two replicates. Cycling conditions were 50°C for two minutes, 95°C for 15 minutes, and 40 cycles of 95°C for 15 sec, 58°C for 30 sec and 72°C for 30 sec.

To avoid the generation of non-specific products, a melting curve analysis of products was routinely undertaken following the amplification. A standard curve was constructed by plotting the cycle number (Ct), at which the amount of target in standard dilutions reaches a fixed threshold, against the log of the amount of starting target. For standard curve construction genomic DNA from the rhesus macaque MMU#13577 was used as a CHEK2 single copy reference. The CHEK2 single copy status in the rhesus macaque genome was verified by both FISH and in silico analysis. Absolute quantification of copy number in the different species was subsequently done by interpolation of the threshold cycle number (Ct) against the corresponding standard curve. Copy numbers of the test genes in primate samples were normalised to the copy number of the CFTR gene (cystic fibrosis transmembrane conductance regulator), which serves as a control representative of a single gene per haploid genome 45 . CFTR primers perfectly match the CFTR gene in all targeted species genomes. The ratio of the CHEK2 copy number to CFTR copy number in each sample normalised the results with respect to differing starting quantity and quality of the template DNA in each reaction 46 . Standard errors of the normalised CHEK2 copy numbers were calculated from the standard deviations of the values of the CFTR and CHEK2 genes using the formula provided by the user menu (ABI Prism 7700 Sequence Detection System, User Bulletin no.2 1997, p.34). Comparisons between the mean values were performed using the Student unpaired t-test. A P-value <0.001 was considered significant.

To investigate the evolution of the CHEK2 duplicon we screened a Y-chromosomal cosmid library (LLOYNC03"M") with a Y-derived CHEK2 duplicon probe (for detail see Methods). A total of 13 positive clones were detected. Probes bordering the Y-chromosomal CHEK2 duplicon were subsequently used to identify cosmids containing the entire derivative duplicon. Out of four positive clones cosmid LLOYNC03"M"22E01 (Figure 1A) was chosen for comparative FISH on human, great ape, Old World monkey (OWM) and New World monkey (NWM) metaphase chromosomes (Figure 1B, Table 1). In the human genome cosmid 22E01 hybridized to euchromatin/heterochromatin transition regions Yq11.1/Yq11.21 and Yq11.23/Yq12, and to additional 12 transition regions on 9 different autosomes. The only interstitial hybridization signal detected was assigned to the ancestral CHEK2 locus in 22q12.1. With only a few exceptions signals were detected in orthologous positions on chimpanzee and gorilla chromosomes (Figure 1B, Table 1). Signals on human and chimpanzee chromosomes 1 and 10 were missing on the orthologous gorilla chromosomes and species-specific signals were restricted to human chromosome 13, and to gorilla chromosomes 2B and 18. In contrast to the multi-site signal patterns found in all African great ape species, in the genomes of orangutan, the rhesus macaque and the common marmoset signals were detected at chromosomal sites orthologous to human 22q12.1. Additional signals with cosmid 22E01 were only detected in the proximal long arm (10qprox) on the NOR (nucleolus organizer region) – containing "marker-chromosome" 10 47 of all three investigated OWM species (rhesus macaque, pig-tailed macaque and baboon). Sequences covered by cosmid 22E01 were subjected to repeat finding programs (for detail see Methods). These analyses suggested that the additional signal on the repeat-rich OWM "marker-chromosome" may be due to diverged CAGGG repeats within the Y-chromosomal cosmid probe. Such CAGGG repeats were previously identified as a single repeat block on the proximal long arm of the "marker-chromosome" of the crab-eating macaque 12 and were shown to have been distributed toward many pericentromeric regions during great ape evolution 12 48 . In addition, preliminary analysis of other Y-chromosomal SDs indicate that there may be further diverged CAGGG repeat sequences located on the Y chromosome that remained undetected by repeat finding programs. These undetected CAGGG repeat sequences may be located in the Yq11.23/Yq12 transition region and in the proximal part of the Yq11.1/Yq11.21 transition region. FISH with large insert genomic clones from both regions showed signals on the proximal long arm of the OWM chromosome 10 25 .

For unambiguous detection of CHEK2 duplicons in anthropoids we performed further FISH experiments with fosmid WI2-1621D20 (G248P81803F6; Figure 2A), which derives from human chromosome 22q12.1. This fosmid contains the complete duplicated portion of the ancestral CHEK2/TTC28 locus. No CAGGG repeat sequences are contained within this CHEK2 duplicon probe. Apart from the signal tagging the ancestral CHEK2 locus in 22q12.1 the probe showed the expected multi-site signal pattern in human (Figure 2B, Table 2). Signals were detected on 8 different chromosomes. On chromosomes 1, 2, 10, 13 and Y signals seemed to be restricted to one chromosomal location, whereas chromosomes 15, 16 and 22 showed a signal cluster, indicating the presence of at least two CHEK2 duplicons on each of these three chromosomes. In contrast to the Y-chromosomal cosmid probe no signals were detected with the fosmid on human chromosomes 9, 14 and Yq11.23/Yq12 (Table 1, Table 2; Figure 1B, Figure 2B). This observation might be explained by the absence of CAGGG repeat sequences in the fosmid probe. The fosmid WI2-1621D20 hybridized to 6 different autosomal chimpanzee chromosomes and the observed signal distribution was highly similar to the human pattern. Nevertheless, no signal was detected on chimpanzee chromosome 13 and only a single signal on chimpanzee chromosome 15. Substantial differences to the Y-chromosomal cosmid probe were detected in the gorilla genome. Only three gorilla chromosomes were tagged. Gorilla chromosome 16 showed two signals located in the proximal and distal transition regions of the heterochromatin block on the short arm (Additional File 1), indicating gorilla-specific reorganization or acquisition of CHEK2 duplicons on this chromosome. Gorilla chromosome 22 showed two distinct signal localizations. One signal mapped to the ancestral interstitial CHEK2 locus and one to the pericentromeric region of the long arm. As in the chimpanzee genome, this probe generated no signal on gorilla chromosome 13, favouring a human-specific CHEK2 duplication event towards the short arm of human chromosome 13. The orangutan genome revealed differences in the signal pattern between both probes, too. In contrast to the Y-chromosomal cosmid probe, two signals were detected on orangutan chromosomes 16 and 22. This observation indicates the occurrence of a first CHEK2 duplication event before the radiation of the great apes and, in addition, shows that sequences located on chromosome 16 are more related to the ancestral than to the derivative locus on the human Y chromosome. Furthermore, hybridization of fosmid WI2-1621D20 on metaphase chromosomes of the white-cheeked crested gibbon, the rhesus macaque, the pig-tailed macaque and the common marmoset revealed a single CHEK2 copy status. In all these species, only the chromosomal region orthologous to human chromosome 22q12 was labelled. This observation contrasts the signal location of the Y-chromosomal cosmid probe in the analysed OWM species, but can again be explained by the absence of CAGGG repeat sequences in the ancestral CHEK2 fosmid probe.

To verify the single-copy status of the non-duplicated portion of the CHEK2 locus, we performed FISH with the fosmid probe WI2-819H21 (G248P81285D11; Figure 2A). In all analysed primate species, including human, this probe hybridized to a single genomic localisation corresponding to the human 22q12.1 orthologous regions. In conclusion, our comparative FISH results with fosmid probe WI2-1621D20 (Figure 2B, Table 2) show the single-copy status of CHEK2 in all tested New World and Old World monkeys and in the white-cheeked crested gibbon. While a first CHEK2 duplication event was detected in the orangutan, a burst of duplication, giving rise to the complex signal pattern, occurred before the radiation of African great apes. Except for chromosomes 16 and Y all signals present in non-human primates were also detected in the orthologous human regions.

We used the sequence from the human ancestral CHEK2 locus for megaBLAST analysis (Basic Local Alignment Search Tool; http://www.ncbi.nlm.nih.gov/blast/Blast.cgi) of the human genome. This analysis revealed the presence of 10 derivative CHEK2 duplicons in the current human reference genome assembly, while two additional duplicons were identified in the Celera whole genome assembly (Figure 3A). The largest CHEK2 duplicon is located in 16p11.2 (16p11.2a) and spans 20747 basepairs of the ancestral CHEK2 locus. This duplicon includes exons 12 to 16 of CHEK2 and exon 1 of TTC28. Furthermore, three additional smaller duplicons (16p11.2b-d) were detected on chromosome 16. Duplicons 16p11.2c and 16p11.2d are about 10 kb in size and share 99% sequence identity, indicating a recent duplication event. In addition, both duplicons terminate within an AluSq-element, which has been shown to be enriched within SD junctions 11 . Slightly shorter than duplicon 16p11.2a is the CHEK2 duplicon located in 15q11.2. A second CHEK2 duplicon on chromosome 15 was only detected in the Celera whole genome assembly and is assigned to chromosome band 15p13. Both chromosome 15 duplicons share 99% sequence identity. As our detailed FISH analysis also indicated the presence of at least two different CHEK2 duplicon copies on human chromosome 15, we conclude that these two duplicons are the result of a recent intrachromosomal duplication event or might reflect individual genomic variation. Furthermore, the 15q11.2 duplicon exhibits an internal 4.2 kb deletion not present in the 15p13 duplicon. Both deletion breakpoints reside in AluSq-elements, one of them being identical to the one terminating the duplicons 16p11.2c and 16p11.2d. This deletion was shown to be present in two different BAC libraries, RPCI-11 and RPCI-13, thus almost certainly ruling out individual variation. In summary, duplicons located at chromosomes 15 and 16 revealed a high intrachromosomal sequence identity. This observation is concordant with previous investigations showing that intrachromosomal duplications share higher sequence identity than interchromosomal duplications, thus pointing to a recent intrachromosomal expansion of the human genome 16 49 . It should be mentioned that the high degree of sequence identity between intrachromosomal duplicons might also be due to nonallelic homologous recombination (NAHR) and/or gene conversion 50 51 52 53 . Additional CHEK2 duplicons, containing almost the entire CHEK2 duplicated portion, are located in 2p11.1/2p11.2, 10p11.1, 22q11.1 and Yq11.1/Yq11.21. Furthermore, one CHEK2 duplicon sequence in the human genome remains chromosomally unassigned (Chr_random), but may be located in the 1q12/1q21 euchromatin/heterochromatin transition region or on the short arm of chromosome 13. FISH signals within these two regions (Figure 2B) do not correspond to any chromosomally assigned CHEK2 duplicon in the current human whole genome assemblies. Interestingly, duplicons located in 2p11.1/2p11.2, 10p11.1, 22q11.1, Yq11.1/Yq11.21 and Chr_random share the same 3 kb LINE1 (long interspersed nuclear element 1) integration, thereby suggesting a common origin. The smallest CHEK2 duplicon was found in 10q11.21 and is about 4.6 kb in size.

To elucidate the molecular evolution of the derivative CHEK2 duplicons we investigated their flanking paralogous sequences. All known CHEK2 duplicons extend into the same proximal duplicated sequence, indicating a single initial pericentromeric "seeding" event. Subsequent pericentromeric "swaps" of the larger duplicon cassette led to the CHEK2 duplicon distribution observed in the human genome. The ancestral location of the duplicon proximal to all CHEK2 duplicons resides within the IGL@ locus in 22q11.21. The duplicated sequences extending beyond the distal CHEK2 duplicon junctions are not identical for all derivative duplicons. While CHEK2 duplicons 16p11.2a and 15q11.2 show the same distal homology, duplicons 2q11.1/2q11.2, 10p11.1, 16p11.2b, 22q11.1, Yq11.1/Yq11.21 and Chr_random share a different distal duplicated sequence. Thus, the latter set of duplicons may have arisen by a duplication of the 16p11.2b duplicon, followed by a LINE1 element integration and successive pericentromeric exchange.

Subsequently, we used the same in silico approach to determine the CHEK2 duplicon architecture of the chimpanzee genome (build2). Chromosomally assigned duplicons in the chimpanzee genome showed chromosomal designations corresponding to the human genome locations. Duplicons were detected on chimpanzee chromosomes 10, 15, 16, 22 and Y (Figure 3B), but not on chromosome 2A. Two duplications containing almost the entire CHEK2 duplicon and 11 smaller duplicon fragments were not chromosomally assigned in the current chimpanzee whole genome assembly. Similar to the human genome the largest CHEK2 duplicon is assigned to chromosome 16 and a slightly shorter CHEK2 duplicon is located on chromosome 15. In contrast to the human genome the chromosome 15 duplicon contains no deletion, indicating that this deletion is human specific. In addition, duplicons on chimpanzee chromosomes 10 and 15 contained small internal duplications and chromosome 16 duplicons seemed to be more fragmented (Figure 3B). These findings might be explained by chimpanzee-specific rearrangements or, more likely, by the inherent problem of generating highly reliable contiguous sequence assemblies in regions enriched in SDs.

Comparative in silico analysis (BLAT search http://genome.ucsc.edu/ of the orangutan (ponAbe2) and rhesus macaque (rheMac2) whole genome assemblies using the human duplicated CHEK2 sequence yield just one copy in the respective genomes on their orthologous chromosome 22. No CHEK2 duplicon was found on orangutan chromosome 16 by this approach, but this may be due to the underrepresentation of segmentally duplicated sequences within pericentromeric regions in the whole genome assemblies 49 . The unduplicated status of CHEK2 in the rhesus macaque genome assembly is concordant with our FISH results obtained with fosmid WI2-1621D20.

To further delineate the evolution of the CHEK2 duplicons we performed phylogenetic analysis with non-coding and non-repetitive sequences using the MEGA4.0 41 software (Figure 4A). The multi-sequence alignment was composed of 1951 basepairs derived from two loci within the CHEK2 duplicons. The proximal sequence is located upstream of exon 1 of TTC28 (NT_011520: 8465388–8465900) and the distal sequence is located in intron 14 of CHEK2 (NT_011520: 8478430–8480219). We aligned all human and chimpanzee derivative duplicons as well as the ancestral sequences of human, chimpanzee and rhesus macaque. Our phylogenetic analyses placed all chromosomally assigned chimpanzee CHEK2 duplicons next to the human CHEK2 duplicons showing the same chromosomal designations (Figure 4A). We therefore concluded the orthologous nature of the particular human and chimpanzee CHEK2 duplicons on chromosomes 10, 15, 16, 22 and Y. Based on an estimated divergence time of 25 million years between human and rhesus macaque lineages 43 , we calculated the effective nucleotide substitution rates (r = K/2 T) (Additional File 2). Calculated times for pericentromeric "seed" and onset of "swaps", are ~22.3 MYA and ~15.9 MYA, respectively. Replacing the human lineage by the chimpanzee lineage, a pericentromeric seed of ~21.2 MYA and an onset of swaps of ~18.0 MYA was calculated. Both independent calculations placed the initial duplication event before the radiation of orangutan and African great apes about 15 MYA 43 thereby coinciding with our FISH results of fosmid clone WI2-1621D20 on orangutan metaphase chromosomes (Table 2; Figure 2B). No signs of duplication were detected on metaphase chromosomes of the white-cheeked crested gibbon, which is thought to have split from the great ape-human lineage about 18 MYA. Only one signal in 7qdist representing the orthologous region of human 22q12 was detected by FISH (Table 2; Figure 2B). This might either be explained by the loss of the primary CHEK2 duplicon in the gibbon lineage or the high degree of sequence divergence between the putative gibbon duplicon and the human ancestral CHEK2 locus. Such a high degree of sequence divergence is a very likely explanation supported by previous investigations, which detected an acceleration of substitution rates after the duplication event 54 55 56 . Phylogenetic reconstruction of the duplication events revealed that the largest human and chimpanzee duplicons located on chromosome 16 share the highest degree of sequence identity with the ancestral CHEK2 locus (Additional File 3). This indicates, in combination with our FISH results, that the first duplication event was a pericentromeric "seed" from the ancestral interstitial CHEK2 locus toward the pericentromeric region of chromosome 16 (Figure 4B). The duplicons located on chromosomes 15 of chimpanzee and human share the highest degree of sequence identity with the largest chromosome 16 duplicons of human and chimpanzee (Additional File 3) suggesting a pericentromeric "swapping" event from chromosome 16 towards chromosome 15. Phylogenetic reconstruction identifies the duplicons located in the pericentromeric regions of human chromosome 2 and human and chimpanzee chromosomes 10, 22 and Y as the most recent ones in the respective genomes. This finding concurs with the in silico results, which showed these duplicons to harbour a unique LINE1 element integration.

The paralogous CHEK2 duplicons in human and African great apes are part of SDs. SDs are known to be frequently associated with copy number variation 57 58 59 60 . A robust approach to specifically target such variable regions is quantitative PCR (qPCR). To independently determine the CHEK2 duplicon numbers in the genomes of anthropoids we performed inter-species quantitative PCR with primers located in exon 14. The complete alignment of exonic sequences from all known CHEK2 duplicons of the human, chimpanzee, orangutan and rhesus macaque genome was used to determine the set of primers best matching the sequences. Nevertheless, we cannot rule out that these primers do not match perfectly to all CHEK2 duplicons in all analysed species. As FISH and in silico analyses consistently demonstrated the single copy status of CHEK2 in the rhesus macaque genome we used rhesus macaque genomic DNA as the reference DNA for CHEK2 copy number prediction in human, great ape and OWM species. We determined the copy numbers of five different human individuals, three chimpanzee and gorilla individuals each, two orangutan and rhesus macaque individuals each and one individual each of the pig-tailed macaque and baboon (Figure 5; Additional File 4). All analysed baboon, pig-tailed macaque, rhesus macaque and orangutan individuals revealed just one CHEK2 copy per haploid genome. A drastic increase in copy number was detected for the African great apes with seven and nine CHEK2 copies in the haploid genomes of gorilla and chimpanzee, respectively. The human genome presented the highest increase in CHEK2 copy number, with a variation between 13 to 16 CHEK2 copies per haploid human genome. Our results significantly demonstrate, that the analysed human individuals have a higher CHEK2 copy number than chimpanzee and gorilla (p < 0.0001) and a higher CHEK2 copy number than orangutan and all analysed OWMs (p < 0.0001). In addition, chimpanzee and gorilla have a significantly higher CHEK2 copy number than orangutan and OWMs (p < 0.0001). This observation confirms our FISH investigation and phylogenetic analysis, which placed the burst of CHEK2 duplication after the radiation of orangutan and African great apes. Furthermore, we detected different numbers of CHEK2 copies within the analysed human individuals. Taking into account the sample size, we can only presume whether there is CHEK2 copy number variation in the human population. As mentioned above, our qPCR approach revealed only one CHEK2 copy per haploid orangutan genome, whereas FISH detected two signals on orangutan chromosomes 16 and 22 (Figure 2B) and phylogenetic analyses indicated the occurrence of a first CHEK2 duplication event before the radiation of orangutan and African great apes (Figure 4). The failure to amplify the second CHEK2 copy in the orangutan genome might be explained either by primer mismatch or loss of exon 14 in the orangutan chromosome 16 copy. Therefore, real CHEK2 copy numbers in non-human primates may be even higher than were predicted by our approach.