Update of the Anopheles gambiaePEST genome assembly
© Sharakhova et al.; licensee BioMed Central Ltd. 2007
Received: 26 July 2006
Accepted: 8 January 2007
Published: 08 January 2007
The genome of Anopheles gambiae, the major vector of malaria, was sequenced and assembled in 2002. This initial genome assembly and analysis made available to the scientific community was complicated by the presence of assembly issues, such as scaffolds with no chromosomal location, no sequence data for the Y chromosome, haplotype polymorphisms resulting in two different genome assemblies in limited regions and contaminating bacterial DNA.
Polytene chromosome in situ hybridization with cDNA clones was used to place 15 unmapped scaffolds (sizes totaling 5.34 Mbp) in the pericentromeric regions of the chromosomes and oriented a further 9 scaffolds. Additional analysis by in situ hybridization of bacterial artificial chromosome (BAC) clones placed 1.32 Mbp (5 scaffolds) in the physical gaps between scaffolds on euchromatic parts of the chromosomes. The Y chromosome sequence information (0.18 Mbp) remains highly incomplete and fragmented among 55 short scaffolds. Analysis of BAC end sequences showed that 22 inter-scaffold gaps were spanned by BAC clones. Unmapped scaffolds were also aligned to the chromosome assemblies in silico, identifying regions totaling 8.18 Mbp (144 scaffolds) that are probably represented in the genome project by two alternative assemblies. An additional 3.53 Mbp of alternative assembly was identified within mapped scaffolds. Scaffolds comprising 1.97 Mbp (679 small scaffolds) were identified as probably derived from contaminating bacterial DNA. In total, about 33% of previously unmapped sequences were placed on the chromosomes.
This study has used new approaches to improve the physical map and assembly of the A. gambiae genome.
The genome of Anopheles gambiae, the major vector of malaria in Africa, was sequenced by a whole-genome shotgun approach . Physical mapping of the genome was conducted by in situ hybridization of about 2,000 bacterial artificial chromosome (BAC) clones on ovarian nurse cell polytene chromosomes. As a result, in the first publication of the A. gambiae physical map, 67 scaffolds equivalent to 227 mega-base-pairs (Mbp) were assigned to chromosomes. Of these, 52 scaffolds were oriented. However, approximately 18% of the assembled A. gambiae genome was represented in scaffolds that did not have a chromosomal location assigned. About 50 Mbp in the assembly were assigned with arbitrary order and orientation to an unmapped chromosome . In this study, new approaches were used to improve the physical map and assembly of the A. gambiae genome.
The most poorly mapped parts of the A. gambiae genome were the pericentromeric regions of the chromosomes. These chromosomal regions are made up of highly and moderately repetitive DNA sequences [3, 4] that are extremely depleted of genes  and form specific heterochromatic structures on chromosomes [6, 7]. Pericentromeric heterochromatin plays an important role in many biological processes, such as cell division , meiotic pairing , regulation of DNA replication and gene expression [10, 11], and is generally associated with gene silencing [12, 13]. However, the assembly and physical mapping of these regions is a difficult part of any genome project [14–19]. In Drosophila melanogaster, for example, one-third of the 180 Mbp genome is centric heterochromatin; but in the first genome publication only 2% of the sequence reads contained heterochromatic simple sequence repeats , and only 3 scaffolds corresponding to 3.8 Mb were mapped in centromeric areas .
According to Cot analysis, 33% (about 86 Mbp) of the A. gambiae genome corresponds to repetitive elements . The highest density of repeats is located in pericentromeric regions and forms the completely heterochromatic Y chromosome . In contrast with Drosophila, short simple repeats are not expanded in the A. gambiae genome; therefore, cloning of the heterochromatic portion of the genome was more successful. However, in the first publication of the A. gambiae genome only 9 scaffolds, with a total size of 3.3 Mbp, were mapped to pericentromeric regions on chromosomes . Mapping is difficult because BAC clones representing pericentromeric regions are likely to map to multiple locations due to their high repeat content. In previous work 27 BAC clones hybridized to all centromeric regions on the chromosomes and 116 BAC clones hybridized to pericentromeric regions and multiple locations on the chromosomes . To determine the genomic location of heterochromatic scaffolds, cDNA clones from the Normalized Anopheles Pool (ANGNAP1) library with sequences matching regions of these scaffolds were mapped to the chromosomes. Additionally, this approach was used to orient scaffolds that were not previously oriented. For some scaffolds, PCR amplified DNA of genes predicted in the scaffolds was used for physical mapping.
No sequence data were assigned to the Y chromosome in the original publication of the A. gambiae genome . Subsequent studies revealed numerous repetitive sequences on the Y chromosome, including four families of satellite DNA and a massive accumulation of several transposable elements, consistent with the fully heterochromatic nature of that chromosome [23, 24]. Only one Y-specific scaffold contained an open reading frame that appeared to correspond to a gene fragment and was expressed exclusively in males. However, a recent extensive bioinformatics-based search failed to reveal other Y-linked scaffolds containing gene sequences .
Another significant problem in the A. gambiae genome assembly was the existence of 64 physical gaps between the mapped scaffolds . BAC and cDNA clones were used for in situ hybridization to physically assign 5 scaffolds with a total size of 1.3 Mbp in these gaps. In addition, systematic in silico analysis of BAC end sequences (BESs) from the ND-1  and ND-TAM  BAC libraries identified BAC clones that span a third of the physical gaps. The sequencing of these clones would allow further improvement of the A. gambiae genome assembly.
Genetic variation within the A. gambiae genome posed another challenge for mapping and assembly . A. gambiae is a highly polymorphic species, characterized by the presence of five chromosomal forms (Bamako, Mopti, Savanna, Bissau and Forest); sympatric populations of the Bamako, Mopti and Savanna forms are at least partially isolated from each other in Mali. These chromosomal forms can be identified by paracentric inversions of the 2R chromosomal arm and have different adaptation to certain climatic conditions and human environments [27–29]. Moreover, an additional type of polymorphism, termed M and S molecular forms, has been revealed in natural populations of A. gambiae by differences in ribosomal DNA [30, 31]. The PEST strain, selected for the genome sequencing because it had the standard chromosomal arrangement, was produced by crossing a laboratory strain originating in Nigeria with the offspring of field-collected A. gambiae from western Kenya . As a result of the high level of polymorphism within the strain, some regions of the genome appeared as two different assemblies ('haplotypes') within the set of scaffolds. Holt et al.  estimated that the presence of alternative assemblies led to overrepresentation of the size of the genome by about 21.3 Mbp. Additional analysis of the scaffold sequences in silico identified 144 previously unplaced scaffolds totaling 8.18 Mbp that are probable alternative assemblies of regions already placed on the chromosomes. In addition, 20 cases totaling 3.53 Mbp of sequence were identified where the adjacent ends of large mapped scaffolds appear to be alternative assemblies of the same region.
The genomic libraries used for the sequencing of the A. gambiae genome were contaminated by bacterial DNA . By bioinformatics approaches, 679 scaffolds with a total size of 1.97 Mbp were determined to be derived from contaminating bacterial DNA.
The revised A. gambiae PEST assembly is available at GenBank. The scaffold entries have information about alternative assembly regions and all other corresponding information. The new RefSeq entries reflect the revised chromosome assemblies (GenBank: CM000356-CM000360).
Physical mapping and scaffold orientation in the pericentromeric regions
Four scaffolds were mapped to the pericentromeric region of chromosome 2 (Additional data file 1). Two scaffolds AAAB01008949 and AAAB01008897 were placed on the right arm (2R) of this chromosome (Figure 2b). Scaffold AAAB01008942 was assigned to the very proximal end of the 2L arm (Figure 2b), and scaffold AAAB01008026 was mapped on the distal part of region 20A (Figure 2b). Both scaffolds mapped to the 2L arm have been oriented. The distal boundary of scaffold AAAB01008987 was also mapped to the telomeric region of the 2R arm. The last BAC clone, 170B21, from this scaffold hybridized to the pair of distal dark bands in subdivision 7A. Additional analysis of previously mapped BAC clones showed that the telomeric end of the 2L chromosomal arm was covered by scaffold AAAB01008807.
Eight scaffolds were assigned to the pericentromeric region of chromosome 3 (Additional data file 1; Figure 2c). Scaffold AAAB01008822 was localized on the distal part of region 37D on the 3R arm. Scaffold AAAB01008943 covered the proximal part of this region and reached the centromeric block of 3R. Scaffolds AAAB01008957, AAAB01008972, AAAB01008985 and AAAB01008981 have been placed in region 38A of the 3L arm. Only two of these scaffolds, AAAB01008943 and AAAB01008972, have been oriented. In addition, scaffold AAAB01008795 was oriented in the region 38C, and the distal boundary of the scaffold AAAB01008849 was more precisely mapped in region 38C (Figure 2c).
The gene content and amounts of transposable elements and short simple repeats were compared between euchromatic and heterochromatic scaffolds across all chromosomes. Gene density in heterochromatin varies but, on average (2 per 100 kbp), is 40% that of the gene density in euchromatin (5 per 100 kbp). In the most centromeric scaffolds, gene content was as low as 0.2 per 100 kbp. The most significant components of the heterochromatic scaffolds are transposable elements: about 50% of the sequences were found to have similarities to the known transposable elements. The content of repeat elements shorter than 200 bp is two-fold higher in heterochromatic scaffolds (8.7%) than in euchromatic scaffolds (4.7%).
Recently, four satellite DNA families were reported from the male-specific Y chromosome ; however, the complete list of scaffolds harboring these satellite sequences was not published. In the present study, 54 such scaffolds have been identified using BLASTN searches. All 54 scaffolds are considered here as Y-linked (Additional data file 2). They usually have short sequences and are composed entirely of repeats of a given satellite family. The few exceptions correspond to scaffolds with juxtaposed arrays of different satellite families or of a satellite DNA and a transposable element fragment. Scaffolds containing Y-linked satellite DNA have a total length of 134 kb. Including the Y-linked scaffold detected previously , the overall length of the Y chromosome scaffolds identified in the A. gambiae genome reaches only 182 kb, making it still the most poorly explored part of the genome. None of the Y-specific scaffolds have been physically mapped, as in situ hybridization experiments were conducted only on polytene chromosomes from ovarian nurse cells.
Assembly improvement in euchromatic regions
Analysis of BAC clones by in situ hybridization was used to assign four additional scaffolds (AAAB01008862, AAAB01008456, AAAB01008882, AAAB01008090) to the euchromatic chromosomal regions (Additional data file 1). In addition, scaffold AAAB01008838 was assigned to an inter-scaffold gap on 3L by in situ hybridization of cDNA clones. Scaffold AAAB01008882 was not included in the final 2L chromosome assembly because of the possibility of some miss-assembly.
Scaffolds from the current Anopheles gambiae genome golden path
Scaffold accession number
Full length of scaffold*
Assembly status to the next scaffold
BAC clones crossing gap between current and next scaffold
Joined and bridged
Joined and bridged
Joined and bridged
Joined and bridged
Joined and bridged
Joined and bridged
Joined and bridged
Joined and bridged
Joined and bridged
04P06, 10C06, 08F18
Joined and bridged
Joined and bridged
Joined and bridged
131N20, 143K17, 23C10
12N18, 08K01, 102F22, 172A24, 19C24, 23K03
Detection of polymorphic and bacterial specific scaffolds
To identify scaffolds from polymorphic regions, unmapped scaffolds were aligned to the chromosome assemblies using the program exonerate , allowing alignments to extend through gaps and possible insertions and deletions in the scaffolds. This revealed 144 scaffolds, of sizes between 15 kbp and 415 kbp and totaling 8.186 Mbp, that aligned over their entire length to a previously mapped, larger scaffold (Additional data file 4). The two aligned alternatives for these regions differed in sequence by between 1.2% and 4.6%, with 90% of the pairs showing sequence differences within the range 1.7% to 3.7%. Such scaffolds probably represent alternative assemblies of the chromosome region and indicate parts of the genome where two haplotypes may have been segregating within the sequenced PEST strain. Seven scaffolds from this list were also physically mapped to appropriate chromosomal locations.
Initial analyses revealed that some of the unmapped scaffolds harbored sequences with unexpectedly high similarity to bacterial proteins. Thirty-two such scaffolds were tested for their presence in the mosquito genome by PCR amplification of A. gambiae PEST strain genomic DNA from embryos using PCR primer pairs specific to these scaffolds (Additional data file 6). Despite repeated attempts, none of the primer pairs yielded any specific products. The combined evidence of high sequence similarity to bacterial genes and negative PCR results strongly suggested that the bacterial-like sequences constitute a contaminant of the A. gambiae genome assembly, rather than an integral part of the A. gambiae genome (data not shown). To identify all such potential bacterial scaffolds, the entire unmapped scaffold set was compared against NCBI's nr protein database. Scaffolds were identified as bacterial contaminants if they had no high similarity to other A. gambiae scaffolds and top hits against the scaffold were only to bacterial proteins with E values at least five orders of magnitude higher than any hits to proteins from eukaryotic organisms. A set of 679 scaffolds, totaling 1.97 Mbp, matched these criteria and are thus regarded as bacterial (Additional data file 6).
The revised assembly (AgamP3) has a total of 80 scaffolds assigned to and ordered on the chromosome arms X, 2R, 2L, 3R and 3L (Table 1). The 28 scaffolds shown in bold have been newly mapped or oriented. In 10 cases, adjacent scaffolds are bridged by BAC clones that have their ends mapped to the two different scaffolds. In 20 cases adjacent scaffolds have been joined because their ends represent alternative assemblies of the same region; 12 of these joins are also supported by bridging BACs. Thus, three different approaches have proved valuable for improving the assembly of the genome: additional physical mapping, detailed in silico analysis of the scaffold sequences, and further mapping of BAC clone end sequences.
This AgamP3 assembly does not complete any of the centromeric regions on the chromosomes, and it is unclear if any of the scaffolds now mapped to centromeric regions actually include functional centromeric sequences. No large blocks of simple repeats appear in the scaffolds that have been mapped in heterochromatic regions. The amount of short repeats (smaller then 200 bp) in different heterochromatic scaffolds varies from 1% to 34%. The functional approximately 420 kbp Drosophila centromere is composed of large blocks of repeats (350 kbp) and more complex sequence composed of transposable elements . The situation is similar in Arabidopsis chromosomes, where the centromeric regions contain tandem 180 bp repeats with a total size of about 0.5 to 3 Mbp, and the surrounding area is enriched in moderate repeats and transposable elements [16, 17]. In neither case does the initial genome assembly reach the centromeric region, and special efforts were required for cloning and sequencing the centromeres.
According to in situ results, the only telomeric region covered by scaffolds in the A. gambiae assembly is on the 2L arm. All three satellite sequences previously described as telomeric [35, 36] have been identified in this scaffold. The in situ results for the distal most BAC clones in the scaffolds closest to the telomeric regions on the other chromosomal arms showed that they are located several bands from the ends of the chromosomes.
The gene content in areas around centromeres is comparable between Anopheles, Drosophila and Arabidopsis genomes. Gene density in the Anopheles genome is 5 per 100 kbp in euchromatic scaffolds, 2 per 100 kbp in pericentromeric and 0.2 per 100 kbp in the three most centromeric scaffolds. In the Drosophila genome the gene content is higher in euchromatin at 11 genes per 100 kbp  and the same at 2 per 100 kbp around the heterochromatin-euchromatin junction . Arabidobsis has an even higher gene content in euchromatic areas of about 25 per 100 kbp, 1.5 in the genetically identified centromeric region and 0.9 in the region enriched in repetitive elements . As in Drosophila  and Arabidopsis , the Anopheles genome does not have a sharp boundary between hetero- and euchromatin.
Figure 4a shows 22 gaps between scaffolds in the A. gambiae genome that can be covered by additional sequencing of BAC clones, which would decrease the number of scaffolds in the genome assembly. The great progress in finishing the Drosophila genome has come as a result of the additional sequencing of overlapping BAC clones, sub-clones and PCR products . Release 3 of the Drosophila genome is represented by 13 scaffolds with a total of 37 sequencing gaps in the euchromatic portion of the genome.
In the initial report of the A. gambiae genomic sequence, Holt et al.  described considerable genetic variation within the PEST strain and suggested, partly on the basis of finding regions with very high single nucleotide polymorphism (SNP) density, that the PEST strain continued to segregate into two different haplotypes for certain regions of the genome. These regions would have derived from the divergent Mopti and Savanna chromosomal forms that contributed to the construction of the PEST strain. Thomasova et al.  sequenced BAC clones in the Pen1 area of the PEST genome and found a 3.3% sequence difference in a 122 kbp region of BAC clone overlap, suggesting that this polymorphism in PEST was not simply an artifact of assembling a highly polymorphic colony. This study has identified 141 distinct scaffolds that probably represent alternative assemblies for regions totaling 8.2 Mbp, and an additional 3.5 Mbp previously mapped to chromosomes. Figure 4a shows the location of the alternative haplotype scaffolds with sizes bigger than 50 kbp. Adjacent pairs of scaffolds that have overlapping alternative assemblies on their ends are shown as single scaffolds on the picture of the new A. gambiae assembly.
It remains possible that some of the sequences designated as alternative haplotype assemblies are actually real duplications. However, the regions identified overlap with those previously found to have high SNP densities, and the alternative assemblies for a region differed in sequence by between 1.2% and 4.6%, similar to the previously reported difference found from BAC sequencing . The identification of scaffolds that represent alternative assemblies enables duplicates to be removed from the set of scaffolds making up the genomic assembly and enables the elimination of artifactual genes from the predicted A. gambiae gene set. It will also facilitate initial studies of both non-coding and gene allelic differences between the two contributing chromosomal forms. It is important to note, however, that the two alternative assemblies of a region are unlikely to accurately represent the two alternative 'haplotypes' that may have been segregating in the PEST strain. Instead, the assembly process may produce two assemblies, both of which are a mosaic of the two haplotypes. Additional scaffolds or scaffold regions that represent alternative assemblies may still be present within the set described here as the revised genomic assembly AgamP3. In this study, 15 kbp was selected as the shortest alignment that could reliably be classified as two alternative assemblies, reasoning that smaller alignments could represent different transposons. In addition, some polymorphic regions may have been assembled as artifactual tandem duplications within a single scaffold ; this study has not attempted to eliminate such regions.
The A. gambiae genomic sequences are expected to contain some level of contamination from bacteria, particularly from those found in the gut . Currently, 679 scaffolds have been identified as apparently bacterial. However, the actual number of bacterial scaffolds within the A. gambiae assembly may be larger. The selected list includes only scaffolds with BLAST hits to bacterial sequences having a cutoff value of E = 10-15. It is likely that some scaffolds with smaller sequence similarity to bacterial sequences currently available in GenBank also have bacterial origin.
Moreover, the assembly refinements described in this paper have a direct impact on the predicted genome wide set of genes. The most recent gene set based on the previous assembly  included 422 gene predictions on scaffolds or scaffold segments now classified as alternative assemblies. The scaffolds now designated as bacterial contaminants had 328 gene predictions already marked as of likely bacterial origin, and an additional 522 not so marked. Inspection of these showed that many had domains suggesting a likely bacterial origin, and none were unequivocally eukaryotic. Hence the first gene set based on the new assembly  benefits from the removal of some duplicate predictions (artifactual paralogues) for genes represented in two alternative assemblies of the same chromosome region and from the absence of predictions derived from bacterial contaminants.
Use of cDNA and BAC clones and PCR amplified gene-fragments as probes for in situ hybridization and additional in silico analysis of the scaffold sequences have led to an overall improvement of the A. gambiae genome assembly. A total of about 15 Mbp has been added to the mapped part of the genome, about 2 Mbp has been removed as bacterial specific and about 12 Mbp has been reclassified as probable alternative assemblies (Figure 4b). One-third of the previously unmapped portion of the A. gambiae genome has been assigned to a chromosomal location. Removal of the probable bacterial and alternative assembly scaffolds has reduced the genome from the original total of 278 Mbp in the Moz2 assembly to 264 Mbp (without gaps), which is much closer to the CoT estimate of 260 Mbp of Besansky and Powell . Moreover, even this new genome size estimate is likely to be somewhat inflated because of residual, small haplotype and bacterial contamination scaffolds. While the AgamP3 assembly can clearly be improved by sequencing BAC clones that cross inter-scaffold gaps or a careful analysis of mate pair violations among the genomic DNA clones sequenced in the original genome project, the upcoming genome projects for the A. gambiae S and M molecular forms  is almost certain to produce a significantly improved assembly for one or both of these two new A. gambiae genomes.
Materials and methods
In situ hybridization
Three different types of probe were used for in situ hybridization: cDNA clones, BAC clones and PCR amplified genes. The cDNA clones were selected from the ANGNAP1 library using BLASTN searching against each of the unmapped and unoriented scaffolds. A pair of cDNA clones from genes on opposite ends of the scaffold and each with a single location in the genome were considered as the candidates for in situ hybridization. BAC clones from ND-1  or ND-Tam  libraries and PCR amplified gene-fragments were identified within the scaffolds using the mappings displayed in the VectorBase genome viewer . Primer design for gene-fragment amplification was done using the Primer 3 program. The cDNA probes were prepared using the FastPlasmid Mini kit (Eppendorf, Hamburg, Germany) and BAC clone DNA was isolated by standard procedures . Both types of probe were labeled with Cy3-AP3-dUTP or Cy5-AP3-dUTP (GE Healthcare UK Ltd (formerly Amersham Biosciences Corp.), Little Chalfont, Buckinghamshire, UK) using the GIBCO BRL Nick Translation labeling system. Amplified gene-fragment DNA was prepared by standard PCR amplification procedures . To prevent non-specific amplification we utilized an appropriate BAC clone DNA as a template for PCR. PCR product was labeled with Cy3-AP3-dUTP or Cy5-AP3-dUTP (GE Healthcare UK Ltd) using the Random Primers DNA Labeling System (Invitrogen, Karlsruht, Germany). To obtain polytene chromosome preparations, ovaries from the SUA strain at the appropriate stage were dissected into fresh Carnoy's solution (ethanol: glacial acetic acid, 3:1). Ovaries were gently pressed with a cover slip in 50% propionic acid, dipped in liquid nitrogen, then cover slips were removed and slides were dehydrated in 50%, 70%, 95%, and 100% ethanol. DNA probes were hybridized to the chromosomes by standard procedures  and then chromosomes were washed in 0.2XSSC, counterstained with YOYO-1 and mounted in DABCO . Fluorescent signals were detected using a Bio-Rad MRC 1024 Scanning Confocal System (Bio-Rad Laboratories, Hercules, CA, USA).
Estimation of gene, transposable element and short repeat density in scaffolds
The genomic sequences of A. gambiae scaffolds were downloaded from the VectorBase website . For the estimation of gene density in scaffolds, 12,600 assembled expressed sequence tags from a Normalized Head library, Normalized Fat Body library and a pooled library were placed on scaffolds using BLAST, requiring an E value of <10-20. For the analysis of the simple repeat content with sizes smaller then 200 bp, we used Tandem Repeats Finder . The percentage of sequences corresponding to the known transposable elements was found using the RepeatMasker program . Two custom databases were used for the search: the database of the A. gambiae transposable elements (Maria Sharakhova and Frank Collins, unpublished data) and the database of natural transposable element sequences identified in D. melanogaster by M Ashburner et al. .
Identification of the Y-specific scaffolds
Scaffolds containing Y chromosome-specific satellite DNA families were regarded as Y-linked. They were identified using randomly selected monomer sequence of each of the four Y-specific satellite DNA families  as a query in local BLASTN searches against the A. gambiae genome database.
Finding BAC clones bridging physical gaps between scaffolds
To identify BAC clones spanning gaps, BLASTN was utilized. A. gambiae BESs that demonstrated significant similarity (E < 10-50) to scaffolds on either side of gaps were selected. From this pool, BACs were identified as crossing gaps if paired BESs fulfilled the following criteria: matching the A. gambiae genomic sequence with an E value of <10-75, having the appropriate relative orientation, and preferably not being repeated on the scaffolds. The sequence length between the BES and the gap was then determined to identify the shortest BAC that crossed the gap, if more than one was identified, using the VectorBase genome viewer.
Identification of polymorphic and bacterial contaminant scaffolds
To detect polymorphic scaffolds in the A. gambiae genome, unmapped scaffolds greater than 15 kbp in length were mapped to the release 2 chromosome assemblies using the program exonerate . Alignment was carried out with exonerate's non-equivalence region (NER) model, which permits alignments to be carried on across the gaps within scaffolds and across possible insertions and deletions. Scaffolds were identified as representing putative alternative assemblies for a region if >97% of the unmapped scaffold was aligned to a chromosome region (where this figure includes any segments treated as non-equivalence regions) and there was >95% sequence identity in the aligned segments. A similar approach was used to identify ends of mapped scaffolds that might represent alternative assemblies of the same region. Selected scaffolds were also aligned and examined using Dotter . Scaffolds were identified as derived from contaminating bacterial DNA if they were unmapped and if the scaffolds appeared to encode only proteins of prokaryotic origin. Scaffold sequences were compared with all proteins in GenBank using BLASTX and were designated as bacterial if they had a hit to a prokaryotic protein with an E value that was <10-15 and 5 orders of magnitude lower than that of the best hit to a eukaryotic protein. To assess the risk of false positive results, we took scaffolds previously mapped to the A. gambiae chromosome arms, broke them into pieces of size equal to the average length of all putative contaminant scaffolds, and then searched them for prokaryotic-like proteins in the same manner. None of the mapped scaffolds would have been designated as bacterial by this procedure.
Additional data files
The following additional data are available with the online version of this paper. Additional data file 1 contains a table listing scaffolds mapped to chromosomes. Additional data file 2 contains a table listing Y chromosome scaffolds. Additional data file 3 contains a table listing BAC clones that span scaffold gaps. Additional data file 4 contains a table containing a list of alternative assemblies. Additional data file 5 contains a table that lists segments of joined scaffolds that represent alternative assemblies of adjacent mapped scaffolds. Additional data file 6 contains a table that lists bacterial specific scaffolds.
We thank Marcelo Bento Soares (University of Iowa) for helping us with construction and normalization of the ANGNAP1 cDNA library. This work was supported by NIAID cooperative agreement U01 AI 48846 and by the NIAID VectorBase Bioinformatics Resource Center contract number HHSN 266200400039C to FHC.
- Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides R, et al: The genome sequence of the malaria mosquito Anopheles gambiae. Science. 2002, 298: 129-149. 10.1126/science.1076181.PubMedView ArticleGoogle Scholar
- Mongin E, Louis C, Holt RA, Birney E, Collins FH: The Anopheles gambiae genome: an update. Trends Parasitol. 2004, 20: 49-52. 10.1016/j.pt.2003.11.003.PubMedView ArticleGoogle Scholar
- John B: From the biology of heterochromatin. Heterochromatin: Molecular and Structural Aspects. Edited by: Verma RS. 1988, Cambridge: Cambridge University Press, 1-147.Google Scholar
- Vaury C, Bucheton A, Pelisson A: The beta heterochromatic sequences flanking the I elements are themselves defective transposable elements. Chromosoma. 1989, 98: 215-224. 10.1007/BF00329686.PubMedView ArticleGoogle Scholar
- Weiler KS, Wakimoto BT: Heterochromatin and gene expression in Drosophila. Annu Rev Genet. 1995, 29: 577-605. 10.1146/annurev.ge.29.120195.003045.PubMedView ArticleGoogle Scholar
- Heitz E: Uber α- und β-Heterochromatin sowie Konstanz und Bau der Chromomeren bei Drosophila. Biol Zentbl. 1934, 54: 588-609.Google Scholar
- Zhimulev IF: Polytene chromosomes, heterochromatin, and position effect variegation. Adv Genet. 1998, 37: 1-566.PubMedView ArticleGoogle Scholar
- Bernard P, Maure JF, Partridge JF, Genier S, Javerzat JP, Allshire RC: Requirement of heterochromatin for cohesion at centromeres. Science. 2001, 294: 2539-2542. 10.1126/science.1064027.PubMedView ArticleGoogle Scholar
- Dernburg AF, Sedat JW, Hawley RS: Direct evidence of a role for heterochromatin in meiotic chromosome segregation. Cell. 1996, 86: 135-146. 10.1016/S0092-8674(00)80084-7.PubMedView ArticleGoogle Scholar
- Swedylow JR, Lamond AI: Nuclear dynamics: where genes are and how they got there. Genome Biol. 2001, 2: REVIEWS0002-Google Scholar
- Marshall WF: Order and disorder in the nucleus. Curr Biol. 2002, 12: R185-192. 10.1016/S0960-9822(02)00724-8.PubMedView ArticleGoogle Scholar
- Henikoff S: Heterochromatin function in complex genomes. Biochim Biophys Acta. 2000, 1470: O1-8.PubMedGoogle Scholar
- Grewal SI, Elgin SC: Heterochromatin: new possibilities for the inheritance of structure. Curr Opin Genet Dev. 2002, 12: 178-187. 10.1016/S0959-437X(02)00284-8.PubMedView ArticleGoogle Scholar
- Hoskins RA, Nelson CR, Berman BP, Laverty TR, George RA, Ciesiolka L, Naeemuddin M, Arenson AD, Durbin J, David RG, et al: A BAC-based physical map of the major autosomes of Drosophila melanogaster. Science. 2000, 287: 2271-2274. 10.1126/science.287.5461.2271.PubMedView ArticleGoogle Scholar
- Carvalho AB, Vibranovski MD, Carlson JW, Celniker SE, Hoskins RA, Rubin GM, Sutton GG, Adams MD, Myers EW, Clark AG: Y chromosome and other heterochromatic sequences of the Drosophila melanogaster genome: how far can we go?. Genetica. 2003, 117: 227-237. 10.1023/A:1022900313650.PubMedView ArticleGoogle Scholar
- Copenhaver GP, Nickel K, Kuromori T, Benito MI, Kaul S, Lin X, Bevan M, Murphy G, Harris B, Parnell LD, et al: Genetic definition and sequence analysis of Arabidopsis centromeres. Science. 1999, 286: 2468-2474. 10.1126/science.286.5449.2468.PubMedView ArticleGoogle Scholar
- Copenhaver GP: Using Arabidopsis to understand centromere function: progress and prospects. Chromosome Res. 2003, 11: 255-262. 10.1023/A:1022887926807.PubMedView ArticleGoogle Scholar
- Horvath JE, Bailey JA, Locke DP, Eichler EE: Lessons from the human genome: transitions between euchromatin and heterochromatin. Hum Mol Genet. 2001, 10: 2215-2223. 10.1093/hmg/10.20.2215.PubMedView ArticleGoogle Scholar
- She X, Horvath JE, Jiang Z, Liu G, Furey TS, Christ L, Clark R, Graves T, Gulden CL, Alkan C, et al: The structure and evolution of centromeric transition regions within the human genome. Nature. 2004, 430: 857-864. 10.1038/nature02806.PubMedView ArticleGoogle Scholar
- Adams MD, Celniker SE, Holt RA, Evans CA, Gocayne JD, Amanatides PG, Scherer SE, Li PW, Hoskins RA, Galle RF, et al: The genome sequence of Drosophila melanogaster. Science. 2000, 287: 2185-2195. 10.1126/science.287.5461.2185.PubMedView ArticleGoogle Scholar
- Besansky NJ, Powell JR: Reassociation kinetics of Anopheles gambiae (Diptera: Culicidae) DNA. J Med Entomol. 1992, 29: 125-128.PubMedView ArticleGoogle Scholar
- Rai KS, Black WC: From mosquito genomes: structure, organization, and evolution. Advances in Genetics. Edited by: Hall JC, Dunlap JC, Friedmann T, Giannell F. 1999, San Diego: Academic Press, a Harcourt Science and Technology Company, 41: 2-33.Google Scholar
- Krzywinski J, Nusskern DR, Kern MK, Besansky NJ: Isolation and characterization of Y chromosome sequences from the African malaria mosquito Anopheles gambiae. Genetics. 2004, 166: 1291-1302. 10.1534/genetics.166.3.1291.PubMedPubMed CentralView ArticleGoogle Scholar
- Krzywinski J, Sangare D, Besansky N: Satellite DNA from the Y chromosome of the malaria vector Anopheles gambiae. Genetics. 2005, 169: 185-196. 10.1534/genetics.104.034264.PubMedPubMed CentralView ArticleGoogle Scholar
- Krzywinski J, Chrystal MA, Besansky NJ: Gene finding on the Y: fruitful strategy in Drosophila does not deliver in Anopheles. Genetica. 2006, 126: 369-375. 10.1007/s10709-005-1985-3.PubMedView ArticleGoogle Scholar
- Hong YS, Hogan JR, Wang X, Sarkar A, Sim C, Loftus BJ, Ren C, Huff ER, Carlile JL, Black K, et al: Construction of a BAC library and generation of BAC end sequence-tagged connectors for genome sequencing of the African malaria mosquito Anopheles gambiae. Mol Genet Genomics. 2003, 268: 720-728.PubMedGoogle Scholar
- Coluzzi M, Sabatini A, Petrarca V, Di Deco MA: Chromosomal differentiation and adaptation to human environments in the Anopheles gambiae complex. Trans R Soc Trop Med Hyg. 1979, 73: 483-497. 10.1016/0035-9203(79)90036-1.PubMedView ArticleGoogle Scholar
- Coluzzi M, Sabatini A, Della Torre A, Di Deco MA, Petrarca V: A polytene chromosome analysis of the Anopheles gambiae species complex. Science. 2002, 298: 1415-1418. 10.1126/science.1077769.PubMedView ArticleGoogle Scholar
- Toure YT, Petrarca V, Traore SF, Coulibaly A, Maiga HM, Sankare O, Sow M, Di Deco MA, Coluzzi M: The distribution and inversion polymorphism of chromosomally recognized taxa of the Anopheles gambiae complex in Mali, West Africa. Parassitologia. 1998, 40: 477-511.PubMedGoogle Scholar
- della Torre A, Fanello C, Akogbeto M, Dossou-yovo J, Favia G, Petrarca V, Coluzzi M: Molecular evidence of incipient speciation within Anopheles gambiae s.s. in West Africa. Insect Mol Biol. 2001, 10: 9-18. 10.1046/j.1365-2583.2001.00235.x.PubMedView ArticleGoogle Scholar
- della Torre A, Tu Z, Petrarca V: On the distribution and genetic differentiation of Anopheles gambiae s.s. molecular forms. Insect Biochem Mol Biol. 2005, 35: 755-769. 10.1016/j.ibmb.2005.02.006.PubMedView ArticleGoogle Scholar
- Slater GS, Birney E: Automated generation of heuristics for biological sequence comparison. BMC Bioinformatics. 2005, 6: 31-. 10.1101/gr.3767105.PubMedPubMed CentralView ArticleGoogle Scholar
- Hoskins RA, Smith CD, Carlson JW, Carvalho AB, Halpern A, Kaminker JS, Kennedy C, Mungall CJ, Sullivan BA, Sutton GG, et al: Heterochromatic sequences in a Drosophila whole-genome shotgun assembly. Genome Biol. 2002, 3: reasearch0085.1-0085.16. 10.1186/gb-2002-3-12-research0085.View ArticleGoogle Scholar
- Sun X, Le HD, Wahlstrom JM, Karpen GH: Sequence analysis of a functional Drosophila centromere. Genome Res. 2003, 13: 182-194. 10.1101/gr.681703.PubMedPubMed CentralView ArticleGoogle Scholar
- Biessmann H, Donath J, Walter MF: Molecularcharacterization of the Anopheles gambiae 2L telomeric region via an integrated transgene. Insect Mol Biol. 1996, 5: 11-20.PubMedView ArticleGoogle Scholar
- Biessmann H, Kobeski F, Walter MF, Kasravi A, Roth CW: DNA organization and length polymorphism at the 2L telomeric region of Anopheles gambiae. Insect Mol Biol. 1998, 7: 83-93. 10.1046/j.1365-2583.1998.71054.x.PubMedView ArticleGoogle Scholar
- Myster SH, Wang F, Cavallo R, Christian W, Bhotika S, Anderson CT, Peifer M: Genetic and bioinformatic analysis of 41C and the 2R heterochromatin of Drosophila melanogaster : a window on the heterochromatin-euchromatin junction. Genetics. 2004, 166: 807-822. 10.1534/genetics.166.2.807.PubMedPubMed CentralView ArticleGoogle Scholar
- Celniker SE, Wheeler DA, Kronmiller B, Carlson JW, Halpern A, Patel S, Adams M, Champe M, Dugan SP, Frise E, et al: Finishing a whole-genome shotgun: release 3 of the Drosophila melanogaster euchromatic genome sequence. Genome Biol. 2002, 3: research0079.1-0079.14. 10.1186/gb-2002-3-12-research0079.View ArticleGoogle Scholar
- Thomasova D, Ton LQ, Copley RR, Zdobnov EM, Wang X, Hong YS, Sim C, Bork P, Kafatos FC, Collins FH: Comparative genomic analysis in the region of a major Plasmodium -refractoriness locus of Anopheles gambiae. Proc Natl Acad Sci USA. 2002, 99: 8179-8184. 10.1073/pnas.082235599.PubMedPubMed CentralView ArticleGoogle Scholar
- Ensembl v20.2b.1 (1 April 2004). [http://ensembl.lcb.uu.se:8080/Anopheles_gambiae/whatsnew/v20_2b_1.html]
- 'VectorBase' Database (Ensembl release v37.3.1). [http://www.vectorbase.org]
- National Human Genome Research Institute: NIH NewsRelease. [http://www.genome.gov/15014493]
- Sambrook J, Frisch EF, Maniatis T: Molecular Cloning: A Laboratory Manual. 1989, New York: Cold Spring Harbor Laboratory Press, 2Google Scholar
- Kumar V, Collins FH: A technique for nucleic acid in situ hybridization to polytene chr of mosquitoes in the Anopheles gambiae complex. Insect Mol Biol. 1994, 3: 41-47.PubMedView ArticleGoogle Scholar
- Sharakhov IV, Serazin AC, Grushko OG, Dana A, Lobo N, Hillenmeyer ME, Westerman R, Romero-Severson J, Costantini C, Sagnon N, et al: Inversions and gene order shuffling in Anopheles gambiae and A. funestus. Science. 2002, 298: 182-185. 10.1126/science.1076803.PubMedView ArticleGoogle Scholar
- Tandem Repeats Finder. [http://tandem.bu.edu/trf/trf.html]
- Institute for System Biology: RepeatMasker. [http://www.repeatmasker.org]
- Berkeley Drosophila Genome Project. [http://www.fruitfly.org]
- Sonnhammer ELL, Durbin R: A dot-matrix program with dynamic threshold control suited for genomic DNA and protein sequence analysis. Gene. 1995, 167: GC1-GC10. 10.1016/0378-1119(95)00714-8.PubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.