The Songbird Neurogenomics (SoNG) Initiative: Community-based tools and strategies for study of brain gene function and evolution
- Kirstin Replogle1, 9,
- Arthur P Arnold2,
- Gregory F Ball3,
- Mark Band8,
- Staffan Bensch11,
- Eliot A Brenowitz4,
- Shu Dong1,
- Jenny Drnevich8,
- Margaret Ferris12,
- Julia M George5,
- George Gong8,
- Dennis Hasselquist11,
- Alvaro G Hernandez8,
- Ryan Kim8,
- Harris A Lewin9, 10,
- Lei Liu8,
- Peter V Lovell6,
- Claudio V Mello6,
- Sara Naurin11,
- Sandra Rodriguez-Zas10,
- Jyothi Thimmapuram8,
- Juli Wade7 and
- David F Clayton1, 9, 12Email author
© Replogle et al; licensee BioMed Central Ltd. 2008
Received: 15 October 2007
Accepted: 18 March 2008
Published: 18 March 2008
Songbirds hold great promise for biomedical, environmental and evolutionary research. A complete draft sequence of the zebra finch genome is imminent, yet a need remains for application of genomic resources within a research community traditionally focused on ethology and neurobiological methods. In response, we developed a core set of genomic tools and a novel collaborative strategy to probe gene expression in diverse songbird species and natural contexts.
We end-sequenced cDNAs from zebra finch brain and incorporated additional sequences from community sources into a database of 86,784 high quality reads. These assembled into 31,658 non-redundant contigs and singletons, which we annotated via BLAST search of chicken and human databases. The results are publicly available in the ESTIMA:Songbird database. We produced a spotted cDNA microarray with 20,160 addresses representing 17,214 non-redundant products of an estimated 11,500–15,000 genes, validating it by analysis of immediate-early gene (zenk) gene activation following song exposure and by demonstrating effective cross hybridization to genomic DNAs of other songbird species in the Passerida Parvorder. Our assembly was also used in the design of the "Lund-zfa" Affymetrix array representing ~22,000 non-redundant sequences. When the two arrays were hybridized to cDNAs from the same set of male and female zebra finch brain samples, both arrays detected a common set of regulated transcripts with a Pearson correlation coefficient of 0.895. To stimulate use of these resources by the songbird research community and to maintain consistent technical standards, we devised a "Community Collaboration" mechanism whereby individual birdsong researchers develop experiments and provide tissues, but a single individual in the community is responsible for all RNA extractions, labelling and microarray hybridizations.
Immediately, these results set the foundation for a coordinated set of 25 planned experiments by 16 research groups probing fundamental links between genome, brain, evolution and behavior in songbirds. Energetic application of genomic resources to research using songbirds should help illuminate how complex neural and behavioral traits emerge and evolve.
Songbirds offer unique opportunities for studying the links between genome, brain structure, neurophysiology, behavior and evolution. They are one of the most successful vertebrate radiations, diversifying into more than 4000 different species over the past ~65 million years [1, 2] and displaying a great range of physical, behavioral and social adaptations [3, 4]. Many songbird species are readily observed in their natural habitats and several species have been domesticated (e.g., zebra finch, canary). All songbirds share a highly specialized neural system for learned vocal communication. Indeed, songbirds are one of the few animal groups capable of complex vocal learning and stand alone as accessible experimental models relevant to human speech [5, 6]. Neurobiological research using songbirds has consistently generated new insights that were later found to be true for other vertebrates; examples include definitive evidence for sexual differentiation of brain circuits, seasonal changes in brain anatomy, and lifelong neuronal replacement .
Against these opportunities, the songbird as a research model also presents challenges. Although the zebra finch has emerged as a primary focus, many other songbird species are studied especially in natural contexts. It will be important to develop tools and reagents that allow study not just of one species (the zebra finch) but of many, so as to exploit the diversity of vocal behavior shown by songbirds. Some of these studies will be comparative in nature (e.g., how is the genome different in species with lifelong vocal learning, compared to species that learn song only once?). Other research objectives may focus on a single wild-caught species to address questions of gene regulation and variation in a particular niche, environment or behavioral paradigm. A related challenge is that researchers with expertise in field biology, neuroscience, physiology or behavioral analyses may not necessarily have equal expertise in molecular genetics, and vice versa.
In response to these needs and opportunities, in 2002 we inaugurated the Songbird Neurogenomics (SoNG) Initiative [8, 9]. Our goals were to leverage rapid advances in genomic technology, so as to bring new resources to songbird research and stimulate collaborative approaches that would integrate field researchers, physiologists and molecular biologists. Here we describe the following accomplishments, which form the foundations of the ongoing SoNG Initiative: 1) three generations in the development of an evolving catalog of gene sequences expressed in the zebra finch brain; 2) production and validation of DNA microarrays useful for study of diverse songbird species; 3) organization of an efficient community-based mechanism for stimulating use of these resources.
High-Throughput Expressed Sequence Analysis and Annotation
Sequence sources and statistics for the three sequential ESTIMA:Songbird assemblies
20 K cDNA microarray
Duke from NCBI
# SEQS INTO PTA
Total unique sequences
Total redundant sequences
ESTIMA:Songbird Build 1
For the primary EST assembly used in production of the SoNG 20 K array (below), cDNA clones were 5'-end-sequenced using high-throughput methods. 35,232 filtered high-quality ESTs were clustered and assembled into 17,878 non-redundant sequences (Table 1), comprising 10,120 pure singlets, 5,855 contigs and an additional 1,903 clustered singlets (i.e., some sequence relationships in initial cluster analysis but could not be assembled into unbroken contigs; possible splice variants). In subsequent EST builds and BLAST analyses (below), 2900 of the Build 1 sequences were absorbed into larger contigs or aligned against redundant gene targets in other species, and 3409 failed to align anywhere (including the whole genome assembly of the chicken). Thus we estimate Build 1 and the corresponding SoNG 20 K microarray represent products coming from between 11,569 and 14,978 unique genes, with the uncertainty depending on the origin of the 3409 ESTs that cannot yet be aligned with a known orthologous genomic region.
ESTIMA:Songbird Build 2
ESTIMA:Songbird Build 2, presence ("yes") or absence ("no") of specific sequences of interest assessed by direct BLAST with available avian or mammalian sequences against 58,211 sequences in the ESTIMA database as of June 2005.
GABA-A alpha 1
estrogen receptor alpha
GABA-A alpha 2
estrogen receptor beta
GABA-A alpha 4
estrogen receptor gamma
GABA-A alpha 5
GABA-A beta 3
GABA-A gamma 2
GABA-A gamma 3/4
Immediate Early Genes
ESTIMA:Songbird Build 3
In a third build of the database, we incorporated two more rounds of subtractive sequencing plus additional sequences from colleagues at Rockefeller University . The combined total of high-quality filtered reads now reached a total of 86,784, which assembled into 31,658 non-redundant transcripts (Table 1). Having reached a marginal redundancy of 64%, we suspended further sequencing of this cDNA source. However, the frequency of singletons remains high (~55%) indicating considerable sequence diversity still remains unrepresented.
Annotation statistics for ESTIMA:Songbird Build 3
20 K array subset
Percent of array
Percent of Build 3 category
Unique sequences after assembly
A) ALL HITS BY DATABASE
Hits against any database
B) Hierarchy of CUSTOM ANNOTATION
1 Use IPI-annotation*
2 in GGA_Unigene but not IPI
3 in HS but not IPI or GGA_unigene
Total number of "custom annotations"
4 additional "conserved in chicken"
Hits against any database
5 remainder = "TGU-specific" (no hits)
*C) IPI Annotations
Number of sequences with IPI identifiers
(Unigene used for 345)
Number of unique IPI identifiers
All identifiers in IPI release 3.26
Fraction of total IPI identifiers
In addition to the 13,219 sequences that aligned with IPI records, another 5614 found matches against the chicken Unigene database but not IPI, and a further 265 aligned against Human Unigene but neither Chicken Unigene nor IPI (Table 3C). In sum, we were able to assign "custom annotation" names for 19,098 sequences, or 60% of the Build 3 assembly. A further 17% showed significant alignment to the chicken genome at sites with no current functional gene annotation, and the remaining 23% have not yet been aligned to any putative orthologous genomic region.
Microarray Production and Validation
Content of SoNG 20 K array
Unique sequences in SB1 assembly
Unique in SB3 assembly
Redundant in SB3 contigs
Controls and knowns
Spots that replicate SB clones
Total DNA spots on array
Blanks and buffer spots
Total addresses on array
Previously Cloned Genes and Replicated Controls spotted on 20 K Array
Seq ID (Gal file)
Replicated control (soybean)
Estrogen Receptor alpha
Estrogen Recetor beta
Wada et al.2
Wada et al.2
Replicated control (soybean)
Retinoic Acid Receptor
Retinoic Acid Receptor
Retinoic Acid Receptor
Replicated control (soybean)
Replicated control (canary "e12" clone)
AMPA subunit, Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
Kainate, Wada et al.2
Wada et al.2
Wada et al.2
Metabotropic, Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
NMDA, Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
Wada et al.2
ZENK cDNA clone ZZF23
ZENK Hybridization Analysis (SoNG 20 K array)
zenk-hybridizing sequences on the 20 K microarray (ordered as in Figure 1).
Canary e12 cDNA
Zebra finch cDNA
Each probe reported an increased signal in the song-stimulated samples versus controls. The redundantly replicated canary zenk control probe measured a two-fold increase in hybridization signal in the song-stimulated samples, somewhat smaller than in past studies where signal for this probe was localized in histological sections by in situ hybridization [20, 21, 29]. Nevertheless the increase was observed here with high statistical confidence (FDR effectively zero). Both the canary and zebra finch control cDNAs reported similar fold-changes and mean signal intensities, suggesting that these hybridization conditions support effective species cross-hybridization for zenk. For the other three EST-derived probes, the magnitudes of increase ranged from 1.2-fold to 2.3-fold. Interestingly, the 5'-most probe measured the smallest fold-change but had the largest mean signal intensity, with the converse for the 3'-most probe. These differences may reflect differential rates of RNA processing along the transcript or sequence relationships to other cross-hybridizing molecules (e.g., zinc finger domains).
Concordance of SoNG 20 K and Lund-zfa arrays
Comparing microarray analyses between different platforms and labs is a difficult task and the results have to be interpreted cautiously. These problems include not only differences in array design and fabrication but also differences in sampling handling [30–32] and differences in the approaches to analyse data from cDNA and short-oligo Affymetrix arrays. Although we acknowledge these problems and the fact that quantitative comparisons are very difficult, we still believe that it is useful to qualitatively compare our two zebra finch array platforms. To the degree the platforms give fundamentally similar results, they validate each other. To compare the SoNG 20 K array and the Lund-zfa array, a set of 11 samples (5 adult males, 6 adult females) were hybridized to both array types.
On the SoNG 20 K array, between 98.4% and 99.8% of the spots on each array (excluding the negative controls) were detectable above local background by the GenePix software, indicating that almost all of the cDNAs on the array are expressed in each of the male and female adult brains (Methods). On the Lund-zfa array, when we used a global background calculated from the cells with the 2% lowest signals + 2 SD we found that 98.7% – 99.9% of the probes were detectable above background (Methods). Since each EST on an Affymetrix array is represented by 11 probes and only 8 probes need to be detectable for significance, virtually all of the ESTs on the arrays will be detected. Hence, both platforms are similar in this rough measure of sensitivity and both detect almost all the transcripts in male and female adult brains.
To compare detection of differentially expressed RNAs, we focused only on those singletons/contigs that could be mapped in a 1:1 relationship across the two platforms, i.e., represented by one clone on the SoNG 20 K array and one probe set on the Lund-zfa array. This resulted in 15,537 comparable gene targets (hereinafter "genes") on each array type (Methods). At a False Discovery Rate (FDR) of 0.05, 492 genes were significantly different between males and females on the SoNG 20 K array, versus 355 genes on the Lund-zfa array. Note that the quantitative differences between the two significant lists are not directly comparable due to differences in sample handling and array analyses. However, despite this the lists shared 286 genes in common, such that most of the expression differences found on the Lund-zfa arrays were also found on the SoNG 20 K array (80.5%). Hence, the fact that so many of the sexually dimorphic genes were identified by both array types shows that the results from the two platforms are in concordance. Presentation and analysis of sex-regulated genes per se is the subject of a separate investigation (A. Arnold and J. Wade, in preparation) to be reported elsewhere.
We conducted a principle components analysis of all 22 sets of array values (using the 15,537 genes in common) after scaling the pre-processed expression values to the overall mean for each gene (Figure 2B). Interestingly, the first principle component (PC1), which explains 16.2 % of the variation, separates the arrays by sex. This shows that the effect of sex is much more important than any variation between platforms that is measured by the other principal components, and that both array platforms are efficient in picking up the biological variation between the samples. The array difference shows up in PC2 (10.3%) & PC3 (8.5%), while the sample pairs do not segregate together until PC4 & PC5 (7.2% and 6.5%). The larger between-sample variation in PC2 (Figure 2B) for the Lund-zfa array compared to the SoNG 20 K array may reflect technical and analytical differences between the hybridizations on the two platforms (for example sample treatment), and does not necessarily indicate an intrinsic difference in the quality of the two arrays.
Comparative genomic hybridization assessment
A major goal of the SoNG Initiative is to facilitate application of genomic analysis to the broad range of songbird species now used in research. To assess the efficacy of our 20 K zebra finch array for measuring gene expression in other songbirds, we performed a set of Comparative Genomic Hybridizations (CGH), where instead of reacting the arrays with labeled tissue mRNA (cDNA), we hybridized with labeled genomic DNA from each of the species. In this way, any differences in signal intensity can be attributed directly to sequence differences between the species, not confounded by differences in gene expression. In general, avian species have a substantially lower repeat density than do mammals, simplifying the analysis of genomic DNA hybridizations [33, 34].
For CGH analysis, we obtained DNA from four other species representative of those proposed for study by the SoNG Initiative collaborators (see next sub-section). White-crowned sparrows and song sparrows are common subjects for field research in North America; the canary is a domesticated species typically studied in the laboratory. Like the zebra finch, all are members of the superfamily Passeroidea. The European starling is studied both in the wild and in captivity. As a member of a different superfamily (superfamily Muscicapoidea), the starling is somewhat more distant from the zebra finch. For comparison we also obtained DNA from two even more distantly related birds, the kingbird and the chicken. The kingbird is a suboscine; like songbirds, suboscines are also members of Order Passeriformes, but suboscines have no apparent vocal learning ability and are estimated to have diverged from oscines (songbirds) about 77 MYA . We also obtained DNA from chickens, a member of a different order of birds that diverged from Passeriformes approximately 100 MYA . Each species was represented by four separate individuals, 2 males and 2 females. Each sample from an individual bird was hybridized on a separate array along with a universal reference sample of genomic DNA pooled from 4 zebra finches (two males and two females) labeled with Cy5. The individual samples were labeled with Cy3 and the universal reference with Cy5 (for our general purpose here we considered the impact of dye bias to be negligible). The resulting data were then expressed as the log(2) ratios of the signal in the two dye channels for each spot on the arrays. A log(2) ratio of zero indicates equal signal from the test species and the reference genomic DNAs.
Interestingly, we also noted cases where particular spots on the zebra finch array gave strongest signals against the genomic DNA of species other than the zebra finch (e.g., negative log values in Fig. 3B). An example is shown in detail in Fig 3D. Note the sharp drop in signal intensity against all species other than zebra finch except for the starling, where a sharp increase in hybridization signal is detected. We speculate that cases like this may represent species differences in gene copy number.
The Community Collaborations
Proposed "Community Collaborations" for songbird microarray studies.
Primary Contrast (question)
Long-tailed, great & penduline tit
House finch, red crossbill
Thrush, Junco, House sparrow, Cordillerian flycatcher
14 non-Tgu species proposed
Using a pipeline for high-throughput sequencing and bioinformatics, we generated ~72,000 reads of sequences expressed in the zebra finch brain. We incorporated these and 27,000 more reads from other projects in the community into a single master database available to the public via the internet . Our approach has been iterative, progressing through three sequential builds each one incorporating additional data. We used our evolving sequence assemblies in the design of two different microarrays for gene expression studies and we have deployed these resources in a large set of collaborations that should continue to bear fruit in the future.
The 86,784 filtered high-quality sequences in the current (third) build of ESTIMA:Songbird describe approximately 31,658 non-redundant transcript segments. Although this is a dramatic increase in the number of sequence information available (there were only 72 records for songbirds in Genbank when our project began ), this collection is still only a partial representation of the sequence diversity in zebra finch brain. Most of the sequences are represented by only a single read and even after three library subtractions, one of every three new sequences generated was still novel. An early analysis of RNA complexity in the canary forebrain suggested a population of as many as 100,000 different transcripts, with total nucleotide diversity approaching the size of the genome itself . This complexity seemed implausible when that analysis was first done, but recent studies in species from flies  to humans [39–41] have now demonstrated that transcription is much more widespread in the genome than previously anticipated.
Some of the 31,658 non-redundant sequences undoubtedly represent discontinuous portions of common transcripts, including alternative splicing products as well as non-overlapping ESTs. Indeed, using the chicken IPI annotation, we detected significant matches for 13,219 sequences but only 8127 (61%) of these IPI identifiers were unique. Presumably this indicates that a substantial fraction of the 31,658 "unique" sequences in the SB3 assembly are in fact derived from a smaller set of genes; by extrapolation this suggests an upper bound of 19,311 (31658 × 61%) unique transcription units represented by the SB3 assembly. Hence we have no reason at this juncture to expect the complement of protein coding genes is any larger for the zebra finch than it is for the chicken, i.e., 20,000–23,000 .
Approximately 23% of ESTIMA:Songbird 3 sequences do not align significantly against the current build of the chicken genome (WASHUC2, May 2006). Some of these could represent orthologs of sequences that are present in the chicken genome but not yet incorporated into the chicken genome build. The current chicken genome assembly is believed to lack approximately 8% of the genes in the full genome complement, including sequences on the female-specific W chromosome in particular. Our cDNA libraries were prepared from a mix of both sexes.
As is true for all gene indices, annotation of function is a major ongoing challenge. Very few genes in songbirds have received any sort of direct experimental analysis thus we must rely on inferences based on similarities to sequences in other species. The closest species with significant genome annotation is the chicken. However, annotation of the chicken genome itself has relied heavily on putative orthologies to mammalian sequences. During our three sequential builds we probed a number of databases to assess their utility for generating putative functional annotations according to sequence similarities. As of Build 3, we focused on the chicken International Protein Index as a primary annotation reference. This is a relatively conservative reference and reflects alignments only with known or predicted protein coding sequences. Only 42% of our Build 3 assembled sequences align against the IPI index, and these represent only a third of the total number of sequence identifiers in the complete chicken IPI index. We anticipate that many of the currently unannotated Build 3 sequences probably represent non-coding segments of IPI-annotated genes. Identification of these relationships should follow from the annotation of the zebra finch genome assembly, expected in 2008.
Our database probably also contains some non-coding RNAs . A preliminary BLAST analysis of the 86,784 ESTs in our current database against miRBASE , a compendium of ~5000 known microRNA sequences, revealed significant alignments for 789 zebra finch ESTs against 489 mature microRNA sequences (data not shown). Additional work will be required to establish whether these sequences are indeed processed into mature microRNAs in the zebra finch.
To stimulate broad application of genomic approaches to songbird research, we generated a cDNA microarray from our first EST build and organized a Community Collaboration system for design and execution of experiments using this array. Part of our group also developed an Affymetrix oligonucleotide array based on our second EST build . To validate these tools and to refine the general methods for the Community Collaborations we did several analyses. These analyses included assessment of optimal methods for RNA purification, MA plots of hybridization data, amplification and labeling from dissected brain samples; and methods for microarray statistical analysis. Some of these studies are summarized in our Call for Community Collaborations (Additional File 2). Validation of the Affymetrix array has been done by the group at Lund University and is not included in this publication. In our main report here, we described several analyses that may have more general implications for transcriptome analysis in songbirds.
The SoNG 20 K cDNA microarray includes 5 different probes for transcripts derived from the zenk gene, including one from the canary sequence. All probes gave signals well above background with labeled cDNA from the dissected auditory lobules of individual birds hearing either song or silence, and all reported an increase in signal in the song-stimulated birds relative to the silence controls. Comparing the near-full positive control probes from zebra finch and canary, both gave equivalent signal intensities and fold-change measurements. In the direct-comparison design used for this experiment, a group size of 6 birds per condition hybridized on 6 arrays was sufficient to reach fairly robust statistical significance for three of the single-spotted probes. The fourth (SB02047B2F12.f1, the 5'-most EST) would not have passed the threshold for identification using even generous criteria (e.g., raw p < 0.05, FDR < 20%). This EST reported a mean intensity that was almost three-fold higher than for the zebra finch and canary positive control probes. We suspect this may indicate cross-hybridization to some other sequences for this particular probe. Alternatively, the variations in mean intensities and fold-changes for the five different probes could indicate differential transcription or RNA processing across the transcript, phenomena that are increasingly observed in high-resolution analyses of transcription units . It is also worth noting that in this preliminary experiment we detected changes in 220 other probes at the significance threshold of the single-spotted zebra finch cDNA postive control (FDR 16%). Analysis of song-regulated sequences is a topic of several of the proposed Community Collaborations and will be described in detail elsewhere.
With the production of the Lund-zfa array after ESTIMA Build 2 , we were able to compare hybridization results using the same tissue samples on the two array platforms. Almost all of the transcripts represented on each platform were detected in adult male and female brains, and there was a general concordance between the platforms on the estimated log2-fold changes for the transcripts they had in common. While the exact "significant" gene lists varied slightly, much of this difference is probably due to differences in sample treatment and analyses between the two array types, which determine the probability of detecting differences. For, example, the SoNG-20 K arrays used a common reference design which helps to normalize differences between arrays, whereas the individual samples had additional shipping and handling before being hybridized to the Lund-zfa arrays, which could have added to their variability. Despite the differences between the platforms, a principle components analysis (Fig. 2B) shows that the biological differences between males and females dominate over any technical differences between the arrays. In sum, both array platforms describe a similar overall picture of differential gene expression and should be useful for further studies.
When we surveyed the songbird research community for interest in using the 20 K array, we recognized that there was considerable interest in experiments involving species other than the zebra finch (Table 7). We initiated comparative genomic hybridizations at this point primarily to evaluate the feasibility of cross-hybridizations with other species. The use of single-species microarrays for species cross-hybridizations is a controversial point in the literature (e.g., pro: [44–50]; con: ; cautionary: [49, 52, 53]). Direct comparison of gene expression in different species on a single-species array is especially problematic, as variations in gene copy number, sequence divergence and RNA expression levels are all confounded. However, for most of the experiments initially proposed for Community Collaborations, the goals were to analyze particular phenomena within a single species, focusing on phenomena that are not well represented in the zebra finch (e.g., effects of photoperiod). We believe that our cross hybridization studies clearly validate our 20 K cDNA microarray for within-species comparisons of other oscine songbirds (Fig. 3), with the caveat that some minority of the probes will give reduced signals compared to hybridizations with zebra finch material. One must also interpret any annotations with particular care, and cross-hybridization results based on particular zebra finch array probes may need to be verified independently, e.g., by resequencing in the target species and RT-PCR to confirm regulation.
The Lund-zfa array was developed with our second build of ESTIMA:Songbird (zebra finch), and it was specifically designed with the intent of supporting research in other songbird species. Quantitative analysis of CGH to DNA from the common whitethroat (Sylvia communis; family Sylviidae) supports the efficacy of cross-hybridization as 96% of the ESTs are called as present . Taking these studies together with our own, zebra finch probes have now been shown to be adequate for detecting signals in the all three major superfamilies of the Passerida parvorder of the oscines (divergence time < 50 MYA) [1, 54, 55]. However, our data for the kingbird (Fig. 3) suggest that use of zebra finch arrays for analysis of sub-oscines (divergence time ~70 MYA) may be more problematic. Direct empirical tests are needed to establish viability for use with Corvid species, oscines that diverged from the Passerida ~50 MYA.
With an inclusive and integrative approach, the Songbird Neurogenomics Initiative has established a rigorous foundation for application of genomic technology by the community of researchers focused on songbirds. Already our ESTIMA:Songbird database has been used in the study of gene dosage compensation in birds vs. mammals , sex chromosome evolution , regulation of brain caspase-3 activity in experience-dependent plasticity , development of markers for study of population genetics in zebra finches , and identification of sequences subject to positive selection in birds (Axelsson et al., submitted). With the third build of the ESTIMA:Songbird database now complete, we anticipate future improvements in transcriptome annotation will soon come from other new approaches including whole genome tiling arrays and massively parallel sequencing [60, 61]. In the meantime, the collection of microarray experiments organized here (Table 7) will generate a broad profile of gene expression in the songbird brain across multiple dimensions of species, sex, neuroanatomy, developmental age, physiological state and behavioral context. The active engagement of a broad research community was instrumental in the selection of the zebra finch as the second avian species for whole genome sequencing, after the chicken . Continuing community participation will be crucial in applying these emerging resources to the expanse of fundamental research questions posed by songbirds.
Sources of cDNA Sequences
Clones for sequencing were drawn first from two normalized libraries (designated SB01 and SB02) representing zebra finch brain RNA from both sexes at multiple ages. Three generations of subtracted libraries (SB03, SB05, SB06) were then prepared sequentially from SB02, by removing the clones sequenced in the previous rounds using PCR subtractive hybridization. For clones from these libraries, sequence reads were generated from the 5' end of each insert. Additional sequence records were also obtained from colleagues at Duke University  and Rockefeller University. Details for each cDNA source follow.
2,880 cDNAs were donated by participant Juli Wade, from a normalized library produced at Michigan State University representing d10-60 male and female telencephalons . Average read length was ~510 nucleotides, and sequence redundancy was 9%.
A new normalized, directional library  was produced for this project using zebra finch brains contributed from aviaries at UCLA (A. Arnold), Oregon Health and Science University (C. V. Mello), Michigan State University (J. Wade) and University of Illinois (D. Clayton). The collected tissues represented telencephalon of both sexes at each of 5 ages: adult, day45, day10, day1 and embryonic. Total RNA was extracted separately from each of the 10 groups, using the Trizol method, and pooled in approximately equal amount. The integrity of total RNA was verified by denaturing agarose gels and by spectrophotometry (ratio A260/280). Poly(A)+mRNA was isolated twice from total RNA using the Oligotex Direct mRNA kit (Qiagen). mRNA was reverse transcribed into double stranded cDNA using a modified oligo18(dT) primer with an identifying tag sequence. Double stranded cDNAs were size selected (more than 450 bp). Size selected cDNAs were adaptored with EcoRI adapters at both ends and digested with NotI. The cDNA was then directionally cloned into EcoR1-NotI digested pBluescript II SK(+) vector (Stratagene). Purified plasmid DNA from the primary library was converted to single-stranded circles and used as a template for PCR amplification using the T7 and T3 priming sites flanking the cloned cDNA inserts. The purified PCR products, representing the entire cloned cDNA population, were used as a driver for normalization. Hybridization between the single-stranded library and the PCR products was carried out for 44 hours at 30°C. Unhybridized single-stranded DNA circles were separated from hybridized DNA rendered partially double-stranded and electroporated into DH10B cells to generate the normalized library. The normalized library had a total of 4.6 × 106 colony forming units (cfu) with less than 1% empty vectors (blue colonies). Average insert size, as determined by PCR of random clones, was 1 kb. Average sequencing read length was ~710 nucleotides. After 18,720 reads, redundancy was 41%.
A subtracted library was produced from SB02 by removing previously sequenced clones as described in . The total number of clones of the SB03 library was 3 × 106 cfu. SB03(a): For ESTIMA:Songbird Build 1, 19,584 clones were sequenced from the subtracted library, with an average clean read length of 700 nucleotides. SB03(b): For ESTIMA:Songbird Build 2, an additional 9,120 sequences were determined.
a second round of subtraction was performed in which clones sequenced from the SB02 and SB03 libraries were removed from the SB03 library as described . The resulting library was named SB05 [there is no SB04 – that designation was used in quality control procedures]. The titer of the subtracted SB05 library was 4.6 × 106. A total of 11,328 clones were sequenced from this library with an average clean read length of 812 nucleotides.
the third round of subtraction involved the removal of all sequenced SB02, SB03 and SB05 clones from the SB05 library. The new SB06 subtracted library had a titer of 2.2 × 106. An additional 11,040 clones were sequenced from SB06, with an average clean read length of 635 nucleotides.
Sequence data for was obtained directly from Erich Jarvis and colleagues at Duke University . Reads had been performed from both 5' and 3' ends of each cDNA. For Build 2, we obtained the raw sequencing chromatograms and processed them as for the SB library reads at Illinois. 64% (23,225) met the minimum quality criteria. For Build 3 we used the final sequence records (17,312) as deposited in NCBI by the Duke investigators, where each record represents a single unconnected read from one end (either 5' or 3') of a single clone. In ESTIMA:Songbird, these sequences are identified by their Genbank IDs, which begin with initial letters "DV".
Sequence data as deposited in NCBI was obtained directly from XiaoChing Li at Rockefeller University . In ESTIMA:Songbird, these sequences are identified by their Genbank IDs, which begin with initial letters "EE".
For high-throughput sequencing from the cDNA libraries described above, bacteria were plated on agar plates with 100 μg/μl of carbenicillin and colonies were robotically picked with the Genetix Q-pix and racked as glycerol stocks in 384-well plates. After overnight growth of the glycerol stocks, bacteria were inoculated into 96-well deep cultures with 0.7 ml of Luria Broth and 100 μg/μl of carbenicillin and grown for 18 hours. Plasmid DNAs were purified from the bacterial cultures with the Qiagen 8000 and Qiagen 9600 BioRobots. Sequencing reactions were set up as follows: 4 μl of 12.5% glycerol, 2 μl of 5X sequencing buffer (Applied Biosystems), 1.5 μl of 1 μM T7 standard primer, 1 μl of H2O, 0.5 μl of BigDye terminator (Applied Biosystems) and 1 μl of template. Thermal cycling was performed at 96°C for 5 min followed by 35 cycles of 96°C for 15 sec, 53°C for 5 sec and 60°C for 4 min. Reaction products were precipitated with 70 μl of 0.2 mM MgSO4 in 70% ethanol. Samples were resuspended in 10 μl of formamide (Applied Biosystems). Sequencing reactions of the 5'ends were performed on ABI 3730 × l capillary systems. All 384- and 96-well format plates were labeled with a barcode and a laboratory information management system (HTLims) was used to track sample flow.
Base-calling was performed using the Phred program. Vector sequences were trimmed using Cross-match. Low-quality bases (Phred quality score < 20) were trimmed from both ends of sequences using Qualtrim and Simpletrim software developed at the W. M. Keck Center (University of Illinois). Those ESTs having a length of more than 200 base pairs after both vector and quality trimming were considered "high-quality" ESTs. The repeat sequences in these ESTs then were masked by the RepeatMasker program using vertebrate repeat sequences database as reference. The masked sequences were then searched for sequences from bacterial chromosomal DNA, RNA, viral DNA, rRNA, and mitochondrial DNA using BLASTN).
Paracel Transcript Assembler (based on CAP4) was used to assemble the ESTs in a two-stage process . First, related sequences were clustered using BLASTN. Each cluster was then assembled into contigs where possible (i.e., no discontinuities in the alignment). All BLAST searches were conducted using the NCBI stand-alone BLAST software. The maximum E-value for an alignment to be judged significant was 10-5, an arbitrary value precedented by other studies [11–13]. The nonredundant EST set was searched by BLASTx against the NCBI nonredundant protein database (NR.aa) and SWISSPROT. Statistics for the three sequential assemblies are given in Table 1. For example, for ESTIMA:Songbird Build 1, 87% of SB01, SB02 and SB03(a) reads passed all criteria above and were included as "filtered, high quality sequences," for a total of 35,232 sequences in the assembly.
BLASTn was used to search against the chicken genome sequence (Gga genome) assembly (WASHUC2, May 2006)  and tBLASTx to align against databases representing chicken transcripts in NCBI (Gga Unigene build 33, NCBI Chicken RNA downloaded May 17, 2007), Ensembl (Chicken cDNA-All and Chicken cDNA ab initio, downloaded May 17, 2007), and the archive  of the TIGR Chicken Gene Index GGGI release 11 . We also used tBLASTx to search against the Human (Hs) Unigene database (Build 201), the NCBI Chicken protein database (downloaded May 17, 2007) and the Chicken International Protein Index (IPI, Release 3.26). Default parameters were used for all BLASTs, which were performed locally with TimeLogic Server. The maximum E-value for an alignment to be judged significant was 10-5, an arbitrary value precedented by other studies [11–13]. Where multiple targets within a database produced alignments with significant E-values, the single best-scoring alignment was saved. Descriptive details for each alignment are available for each record in the ESTIMA:Songbird database . To obtain a short, descriptive working name for annotated sequences, where available we imported the IPI protein name into our ESTIMA database in the "custom annotation" field.
Production of 20 K spotted cDNA Array
Individual cDNA clones were selected to represent each of the 17,878 unique sequences in the primary assembly. For contigs that comprised multiple clones, we selected a single clone for the microarray as follows:
a. for each EST, we compared top 5 hits against NCBI(NR), vs top 5 for whole contig consensus sequence; eliminated ESTs having no correspondence to any of the contig's hits [rationale: their inclusion in the contig could be an artifact – direct similarity to the core conserved consensus is validating].
b. selected the EST farthest upstream ["5'-most," should be the longest, and the most useful for cross-hybridizations]
c. where multiple ESTs shared the 5'-most ends, we picked the longest.
d. for the minority of contigs that did not align against NR, we found the longest ORF in the contig consensus and selected the EST that gave the best match (by TBLASTN) to the consensus ORF.
The cDNAs representing these clones and controls were rearrayed in 384 well plates with a Genetix QPix clonepicking instrument and stored as a stock clone set at -80°C. Sample clones from each plate were resequenced for quality control and accuracy of clone transfer. Stock clones were inoculated for overnight cultures in LB media containing 75 μg/ml ampicillin and 8% glycerol. 7.5 μl diluted (1:14) culture was added to 75 μl PCR reactions containing 37.5 μl custom 2X Platinum PCR supermix (Invitrogen, Carlsbad, CA), 0.3 μl of 100 μM forward and reverse primers (representing the vector sequence flanking the cDNA inserts). PCR parameters were as follows: Initial denaturation at 95°C, 10 min. followed by 35 cycles of 95°C, 40 sec., 65°C, 40 sec, 72°C, 3 min, and a final extension at 72°C for 5 min. PCR products were checked by electrophoresis on E-gel 96 1% gels (Invitrogen, Carlsbad, CA). Failed reactions were reamplified as above with the addition of 1 M Betaine to the PCR reaction. PCR product was cleaned with Millipore PCR 96 cleanup plates (Millipore, Billerica, MA). All liquid manipulations were carried out in 96 well format on a Beckman Biomek FX (Beckman Coulter, Fullerton, CA). Cleaned product was transferred to 384 well plates, dried in a speedvac and resuspended in 3 × SSC and 1.5 M Betaine buffer for printing. using a GeneMachines OmniGrid 100 robotic printer onto Corning GAPSII amino-silane coated slides. PCR product sufficient for printing 1200 arrays was generated during July-September, 2004.
All procedures involving animals were conducted with formal institutional (IACUC) approval and oversight. For the microarray analysis, male zebra finches 3–17 months old (age-balanced, two groups) were hatched and reared (3 birds/cage; 12:12 light:dark cycle) in D. Clayton's aviary at the Beckman Institute, Urbana IL. During 2/20/2005–3/10/2005, each was individually isolated in a sound attenuation chamber for 46 hours, then exposed to a 30 minute song playback (or silence) between noon and 1 PM. Behavioral responses were videotaped for subsequent quantification of a "listening" index . At the end of the 30 minute stimulation period, the birds were immediately euthanized by decapitation, and the two auditory lobules (AL) were dissected out , put into RNAse-free 1.5 ml tubes and frozen on a dry ice/ethanol mixture. Tissue was stored at -80°C until RNA preparation.
Microarray hybridization (Song stimulation experiment)
Total RNA was prepared with RNAqueous-Micro kit (Ambion; average yield = 4 μg/AL-pair). 500 ng RNA was amplified using the Low RNA Input Fluorescent Linear Amplification kit (Agilent; average yield = 25 μg). The resulting aRNA was reverse transcribed using an indirect aminoallyl incorporation protocol and labeled with either Cy3or Cy5 dyes (GE Healthcare). Dye labeling was balanced by group (i.e., half of each group, song or silence, was labeled with Cy3 and the other half with Cy5). Six arrays were hybridized simultaneously, each array with one Song and one Silence sample. Thus each array represents one of six biological replicate measurements of the song-vs-silence relationship for each spot on the array. Slides were hybridized overnight at 42°C, washed and scanned using an Axon GenePix 4000B microarray scanner. All slide images were analyzed using GenePix Pro 6.0 software. Analyzed slide images were manually edited and aberrant spots were flagged for exclusion in downstream analysis.
Statistical analysis of hybridization data (Song stimulation experiment)
Prior to analysis, the fluorescence intensities were edited by removing automatically and manually flagged spots that did not surpass minimum quality thresholds. "Background" was defined independently for each spot from its surround and subtracted from each spot value. The data were log2 transformed and loess normalized within each array. Analysis then proceeded using the two-step approach of Cui et al. , where the normalized data were first adjusted for global array and dye effects across genes, then the resulting adjusted intensities were analyzed by gene with a parametric linear model [73–76]. This model was implemented using Bioconductor R-routines  and the MIXED procedure (SAS, 2004).
For the analysis of the five different zenk sequences (Figure 1, Table 6), replicate spottings of the same cDNA on one array (e.g., the redundant spottings of the canary zenk cDNA) were treated as replicate measurements, whereas cDNAs representing different segments of the same gene were treated as independent targets (e.g, each of the other four zebra finch zenk cDNAs spotted once per array).
Production of Universal Reference RNA sample
Total RNA was prepared from whole telencephalon of 30 birds [3 males and 3 females from each of the aviaries of Clayton, Wade, Arnold, Mello, and D. Perkel (University of Washington)] using TRI Reagent (Ambion). The total RNA was DNase I treated (Turbo DNase, Ambion) and cleaned up on a spin column (RNeasy, QIAgen). Equal amounts of RNA from each bird were pooled. Aliquots of the pool were amplified (Low Input Linear RNA Amplification Kit, Agilent) and pooled. Sufficient cRNA for 5× more arrays than necessary was generated and stored at -80 in small aliquots. A test hybridization of the reference sample, labeled only with Cy3 (GE Healthcare), showed signal detectable above background on greater than 95% of probes.
Concordance of SoNG 20 K and Lund-zfa arrays
RNA samples were prepared from 11 individual zebra finch brains (5 adult males, 6 adult females) and hybridized to SoNG 20 K arrays as described above for the Song Stimulation experiment, except that a common reference design was used. One male or female sample was hybridized to one channel of an array and the universal reference sample was hybridized to the other channel of each array. No technical dye-swaps were done; instead possible dye-bias was controlled by putting the reference sample in the Cy3 channel for ~ half the male and ~ half the female samples and in the Cy5 channel for the other half of the samples. After the analysis in the US the samples were kept at -80°C for 2 months before they were shipped on dry ice to an Affymetrix service provider in Sweden, the Swegene Center for Integrative Biology at Lund University (SCIBLU MARC) , where hybridizations to the Lund-zfa array were performed. The experiments in the present study used the initial version of the Lund-zfa array which was an Affymetrix design spotted with NimbleGen technology . (Note that since then, the Lund-zfa Array has been re-designed and the version currently distributed is spotted like any Affymetrix standard array.). Before hybridization the RNA samples were quality checked using a Nanodrop spectrophotometer and RIN values were calculated using an Agilent 2100 Bioanalyzer. One high quality sample representing each of the 11 individuals was hybridized according to standard Affymetrix protocols for RNA. Five μg total RNA from each sample was used in the regular protocols for GeneChip® Arrays and hybridized onto the Lund-zfa Affymetrix array overnight in the GeneChip® Hybridisation oven 6400 using standard procedures. The arrays were washed and then stained in a GeneChip® Fluidics Station 450. Scanning was carried out with the GeneChip® Scanner 3000 and image analysis was performed using GeneChip® Operating Software.
Pre-processing of each array type was done in R using Bioconductor packages appropriate for spotted or Affymetrix arrays. Because > 98% of spots on the SoNG 20 K arrays were above local background estimates, no background correction was performed. Within-array printtip loess normalization was performed, as was a between-array scaling normalization using the limma package . For the Lund-zfa arrays, pre-processing was done using RMA (with the normal background correction included) . The Lund-zfa array is a PM only array, with Perfect Match probes arranged in a checkerboard pattern on the array and 4 empty features adjacent to each PM probe . We used the signals from the 2% lowest empty features on the chip (5111 features distributed throughout the spotted area) + 2 standard deviations as an estimate of the global background of each chip, and this background cut-off was then used to calculated the percentage of present call probes. To use the 2% lowest signals is a standard Affymetrix cut-off for background calculation. Both array types were analyzed for differential expression using the limma package, which fits a mixed linear model with an empirical Bayes error correction. Multiple test correction of the p values was done using the false discovery rate (FDR) method .
Comparative Genomic Hybridization (CGH)
Tissue or blood was obtained from two males and two females of each species. Genomic DNA (gDNA) was collected from each sample using the GenomicPrep Cell and Tissue or Blood DNA Isolation Kit (GE Healthcare). gDNA was digested with Hae III for 5 hours at 37°C and cleaned using a Qiaquick PCR purification kit (QIAgen). A reference sample was created by pooling equal amounts of digested and cleaned DNA from 2 male and 2 female zebra finches. Template DNA was polymerized from 4 μg digested and cleaned DNA using high concentration Klenow (New England Biolabs) and incorporating amino-allyl dUTP (Sigma Aldrich) for 4 hours at 37°C. The resulting amino-allyl labeled DNA was coupled with Cy dye (GE Healthcare), cleaned, and hybridized using SlideHyb #1 hybridization buffer (Ambion). All experimental samples were labeled with Cy3 and all reference samples were labeled with Cy5. Slides were hybridized for 48 hrs at 42°C, washed and scanned using an Axon GenePix 4000B microarray scanner. All slide images were analyzed using GenePix Pro 6.0 software. Analyzed slide images were manually edited and aberrant spots were flagged for exclusion in downstream analysis. Normalization and statistical analysis  was performed using the TM4 software suite developed by TIGR. Slides were within-slide LOWESS normalized using MIDAs.
All derived sequence resources are publicly accessible via internet at the Songbird Neurogenomics website , which includes a link to the ESTIMA:Songbird EST database . Resources include raw sequencing chromatograms, contig alignments, text files of both raw and trimmed sequence, BLAST searching of the database, a GO Browser, and annotations imported from BLAST searches against the databases listed in Table 3. All ESTs have been deposited in Genbank (Accession numbers DV944971 - DV962014, CK301200 - CK317559 and FE712085 - FE739917). It should be noted that the Genbank records are restricted to "high quality" trimmed sequence data, which typically represents only a portion of each cDNA spotted on the array. For access to the Duke-derived annotations of their cDNAs, we incorporated direct links to the Duke website  in the records presented on the ESTIMA:Songbird website. Clones in the ESTIMA database can be purchased through the Clemson University Genomics Institute (CUGI)  distribution service. Raw microarray data is archived and distributed in MIAME-compliant form using 3rd Millenium's ARDAS (Array Repository and Data Analysis System) server  and is accessible using "public" (no quotes) as the login and password.
Comparative Genomic Hybridization
auditory lobule of the caudomedial telencephalon
Gambel's white-crowned sparrow (Zonotrichia leucophry gambelii)
Song sparrow (Melospiza melodia)
Western Kingbird (Tyrannus verticalis)
Canary (Serinus canaria)
European starling (Sturnus vulgaris)
Chicken (Gallus gallus)
Gallus gallus (chicken)
Taeniopygia guttata (zebra finch)
Expressed Seqence Tag
Statistical Analysis of Microarrays
Million Years Ago
False Discovery Rate
University of Illinois at Urbana-Champaign
Supported by NIH grant NS045264 and the Swedish Research Council (grants to SB and DH). The authors thank Wesley Warren and Gene Robinson for many helpful discussions; the Burke Museum of the University of Washington for providing Tyrannus verticalis tissue (specimen numbers UWBM 77787, UWBM 77788, UWBM 77789, UWBM 78000); Lila Vodkin for providing soybean control genes; and Fiona McCarthy for advice on functional annotation.
- Barker FK, Cibois A, Schikler P, Feinstein J, Cracraft J: Phylogeny and diversification of the largest avian radiation. Proc Natl Acad Sci USA. 2004, 101: 11040-11045. 10.1073/pnas.0401892101.PubMedPubMed CentralView Article
- Ericson PG, Anderson CL, Britton T, Elzanowski A, Johansson US, Kallersjo M, Ohlson JI, Parsons TJ, Zuccon D, Mayr G: Diversification of Neoaves: integration of molecular sequence data and fossils. Biol Lett. 2006, 2: 543-547. 10.1098/rsbl.2006.0523.PubMedPubMed CentralView Article
- Baptista LF, Trail PW: The role of song in the evolution of passerine diversity. Syst Biol. 1992, 41: 242-247. 10.2307/2992524.View Article
- Orenstein R: Encyclopedia of Songbirds. Toronto: Key Porter Books
- Doupe AJ, Kuhl PK: Birdsong and human speech: Common themes and mechanisms. Annual Review of Neuroscience. 1999, 22: 567-631. 10.1146/annurev.neuro.22.1.567.PubMedView Article
- Kuhl PK: Human speech and birdsong: communication and the social brain. Proc Natl Acad Sci USA. 2003, 100: 9645-9646. 10.1073/pnas.1733998100.PubMedPubMed CentralView Article
- Zeigler HP, Marler P: Behavioral Neurobiology Of Bird Song. 2004, Annals New York Acad Sci, 1016: 1-778.
- Songbird Neurogenomics Initiative. [http://titan.biotec.uiuc.edu/songbird/]
- ESTIMA:Songbird. [http://www.uiuc.edu/goto/songbird]
- Kumar CG, LeDuc R, Gong G, Roinishivili L, Lewin HA, Liu L: ESTIMA, a tool for EST management in a multi-project environment. BMC Bioinformatics. 2004, 5: 176-10.1186/1471-2105-5-176.PubMedPubMed CentralView Article
- Band MR, Olmstead C, Everts RE, Liu ZL, Lewin HA: A 3800 gene microarray for cattle functional genomics: comparison of gene expression in spleen, placenta and brain. Anim Biotech. 2002, 163-172. 10.1081/ABIO-120005779.
- Whitfield CW, Band MR, Bonaldo MF, Kumar CG, Liu L, Pardinas JR, Robertson HM, Soares BE, Robinson GE: Annotated expressed sequence tags and cDNA microarrays for studies of brain and behavior in the honey bee. Genome Res. 2002, 12: 555-566. 10.1101/gr.5302.PubMedPubMed CentralView Article
- W.M. Keck Center for Comparative and Functional Genomics, UIUC. [http://www.biotech.uiuc.edu/centers/Keck/]
- Wada K, Howard JT, McConnell P, Whitney O, Lints T, Rivas MV, Horita H, Patterson MA, White SA, Scharff C, Haesler S, Zhao S, Sakaguchi H, Hagiwara M, Shiraki T, Hirozane-Kishikawa T, Skene P, Hayashizaki Y, Carninci P, Jarvis ED: A molecular neuroethological approach for identifying and characterizing a cascade of behaviorally regulated genes. Proc Natl Acad Sci USA. 2006, 103: 15212-15217. 10.1073/pnas.0607098103.PubMedPubMed CentralView Article
- Naurin S, Bensch S, Hansson B, Johansson T, Clayton DF, Albrekt AS, Torbjörn von Schantz T, Hasselquist D: A microarray for large-scale genomic and transcriptional analyses of the zebra finch (Taenopygia guttata) and other passerines. Mol Ecol Resources. 8 (2): 275-281. 10.1111/j.1471-8286.2007.01979.x.
- Li X, Wang XJ, Tannenhauser J, Podell S, Mukherjee P, Hertel M, Biane J, Masuda S, Nottebohm F, Gaasterland T: Genomic resources for songbird research and their use in characterizing gene expression during brain development. Proc Natl Acad Sci USA. 2007, 104: 6834-6839. 10.1073/pnas.0701619104.PubMedPubMed CentralView Article
- Kersey PJ, Duarte J, Williams A, Karavidopoulou Y, Birney E, Apweiler R: The International Protein Index: an integrated database for proteomics experiments. Proteomics. 2004, 4: 1985-1988. 10.1002/pmic.200300721.PubMedView Article
- Nastiuk KL, Mello CV, George JM, Clayton DF: Immediate-early gene responses in the avian song control system: Cloning and expression analysis of the canary c-jun cDNA. Mol Brain Res. 1994, 27: 299-309. 10.1016/0169-328X(94)90013-2.PubMedView Article
- Velho TA, Pinaud R, Rodrigues PV, Mello CV: Co-induction of activity-dependent genes in songbirds. Eur J Neurosci. 2005, 22: 1667-1678. 10.1111/j.1460-9568.2005.04369.x.PubMedView Article
- Mello CV, Vicario DS, Clayton DF: Song presentation induces gene expression in the songbird forebrain. Proc Natl Acad Sci USA. 1992, 89: 6818-6822. 10.1073/pnas.89.15.6818.PubMedPubMed CentralView Article
- Mello CV, Nottebohm F, Clayton DF: Repeated exposure to one song leads to a rapid and persistent decline in an immediate early gene's response to that song in zebra finch telencephalon. J Neurosci. 1995, 15: 6919-6925.PubMed
- Kruse AA, Stripling R, Clayton DF: Context-specific habituation of the zenk gene response to song in adult zebra finches. Neurobiol Learn Mem. 2004, 82: 99-108. 10.1016/j.nlm.2004.05.001.PubMedView Article
- Jin H, Clayton DF: Localized changes in immediate-early gene regulation during sensory and motor learning in zebra finches. Neuron. 1997, 19: 1049-1059. 10.1016/S0896-6273(00)80396-7.PubMedView Article
- Jarvis ED, Scharff C, Grossman MR, Ramos JA, Nottebohm F: For whom the bird sings – context-dependent gene expression. Neuron. 1998, 21: 775-788. 10.1016/S0896-6273(00)80594-2.PubMedView Article
- Bailey DJ, Wade J: Differential expression of the immediate early genes FOS and ZENK following auditory stimulation in the juvenile male and female zebra finch. Brain Res Mol Brain Res. 2003, 116: 147-154. 10.1016/S0169-328X(03)00288-2.PubMedView Article
- Bailey DJ, Wade J: FOS and ZENK responses in 45-day-old zebra finches vary with auditory stimulus and brain region, but not sex. Behav Brain Res. 2005, 162: 108-115. 10.1016/j.bbr.2005.03.016.PubMedView Article
- Bolhuis JJ, Hetebrij E, Den Boer-Visser AM, De Groot JH, Zijlstra GG: Localized immediate early gene expression related to the strength of song learning in socially reared zebra finches. Eur J Neurosci. 2001, 13: 2165-2170. 10.1046/j.0953-816x.2001.01588.x.PubMedView Article
- Terpstra NJ, Bolhuis JJ, Riebel K, van der Burg JM, den Boer-Visser AM: Localized brain activation specific to auditory memory in a female songbird. J Comp Neurol. 2006, 494: 784-791. 10.1002/cne.20831.PubMedView Article
- Mello CV, Clayton DF: Song-induced ZENK gene expression in auditory pathways of songbird brain and its relation to the song control system. J Neurosci. 1994, 14: 6652-6666.PubMed
- Shi L, Reid LH, Jones WD, Shippy R, Warrington JA, Baker SC, Collins PJ, de Longueville F, Kawasaki ES, Lee KY, Luo Y, Sun YA, Willey JM, Setterquist RA, Fischer GM, Tong W, Dragan YP, Dix DJ, Frueh FW, Goodsaid FM, Herman D, Jensen RV, Johnson CD, Lobenhofer EK, Puri RK, Schrf U, Thierry-Mieg J, Wang C, Wilson M, Wolber PK, Zhang L, Slikker WJ, Shi L, Reid LH: The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat Biotech. 2006, 24: 1151-1161. 10.1038/nbt1239.View Article
- Irizarry RA, Warren D, Spencer F, Kim IF, Biswal S, Frank BC, Gabrielson E, Garcia JG, Geoghegan J, Germino G, Griffin C, Hilmer SC, Hoffman E, Jedlicka AE, Kawasaki E, Martinez-Murillo F, Morsberger L, Lee H, Petersen D, Quackenbush J, Scott A, Wilson M, Yang Y, Ye SQ, Yu W: Multiple-laboratory comparison of microarray platforms. Nat Methods. 2005, 2: 345-350. 10.1038/nmeth756.PubMedView Article
- Guo L, Lobenhofer EK, Wang C, Shippy R, Harris SC, Zhang L, Mei N, Chen T, Herman D, Goodsaid FM, Hurban P, Phillips KL, Xu J, Deng X, Sun YA, Tong W, Dragan YP, Shi L: Rat toxicogenomic study reveals analytical consistency across microarray platforms. Nat Biotech. 2006, 24: 1162-1169. 10.1038/nbt1238.View Article
- Hillier LW, Miller W, Birney E, Warren W, Hardison RC, Ponting CP, Bork P, Burt DW, Groenen MA, Delany ME, Dodgson JB, Chinwalla AT, Cliften PF, Clifton SW, Delehaunty KD, Fronick C, Fulton RS, Graves TA, Kremitzki C, Layman D, Magrini V, McPherson JD, Miner TL, Minx P, Nash WE, Nhan MN, Nelson JO, Oddy LG, Pohl CS, Randall-Maher J, Smith SM, Wallis JW, Yang SP, Romanov MN, Rondelli CM, Paton B, Smith J, Morrice D, Daniels L, Tempest HG, Robertson L, Masabanda JS, Griffin DK, Vignal A, Fillon V, Jacobbson L, Kerje S, Andersson L, Crooijmans RP, Aerts J, van der Poel JJ, Ellegren H, Caldwell RB, Hubbard SJ, Grafham DV, Kierzek AM, McLaren SR, Overton IM, Arakawa H, Beattie KJ, Bezzubov Y, Boardman PE, Bonfield JK, Croning MD, Davies RM, Francis MD, Humphray SJ, Scott CE, Taylor RG, Tickle C, Brown WR, Rogers J, Buerstedde JM, Wilson SA, Stubbs L, Ovcharenko I, Gordon L, Lucas S, Miller MM, Inoko H, Shiina T, Kaufman J, Salomonsen J, Skjoedt K, Wong GK, Wang J, Liu B, Yu J, Yang H, Nefedov M, Koriabine M, Dejong PJ, Goodstadt L, Webber C, Dickens NJ, Letunic I, Suyama M, Torrents D, von Mering C, Zdobnov EM, et al: Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature. 2004, 432: 695-716. 10.1038/nature03154.View Article
- Ellegren H: The avian genome uncovered. Trends Ecol Evol. 2005, 20: 180-186. 10.1016/j.tree.2005.01.015.PubMedView Article
- van Tuinen M, Hedges SB: Calibration of avian molecular clocks. Mol Biol Evol. 2001, 18: 206-213.PubMedView Article
- Clayton DF: Songbird genomics: methods, mechanisms, opportunities, and pitfalls. Annals New York Acad Sci. 2004, 1016: 45-60. 10.1196/annals.1298.028.View Article
- Clayton DF, Huecas M: Forebrain-enriched RNAs of the canary: a population analysis using hybridization kinetics. Mol Brain Res. 1990, 7: 23-30. 10.1016/0169-328X(90)90069-P.PubMedView Article
- Manak JR, Dike S, Sementchenko V, Kapranov P, Biemar F, Long J, Cheng J, Bell I, Ghosh S, Piccolboni A, Gingeras TR: Biological function of unannotated transcription during the early development of Drosophila melanogaster. Nat Genet. 2006
- Peters BA, St Croix B, Sjoblom T, Cummins JM, Silliman N, Ptak J, Saha S, Kinzler KW, Hatzis C, Velculescu VE: Large-scale identification of novel transcripts in the human genome. Genome Res. 2007, 17: 287-292. 10.1101/gr.5486607.PubMedPubMed CentralView Article
- Consortium ENCODEP: Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature. 2007, 447: 799-816. 10.1038/nature05874.View Article
- Johnson JM, Edwards S, Shoemaker D, Schadt EE: Dark matter in the genome: evidence of widespread transcription detected by microarray tiling experiments. Trends Genet. 2005, 21: 93-102. 10.1016/j.tig.2004.12.009.PubMedView Article
- Eddy SR: Noncoding RNA genes. Curr Opin Genet Dev. 1999, 9: 695-699. 10.1016/S0959-437X(99)00022-2.PubMedView Article
- Griffiths-Jones S, Grocock RJ, van Dongen S, Bateman A, Enright AJ: miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res. 2006, 34: D140-4. 10.1093/nar/gkj112.PubMedPubMed CentralView Article
- Pollack JR, Perou CM, Alizadeh AA, Eisen MB, Pergamenschikov A, Williams CF, Jeffrey SS, Botstein D, Brown PO: Genome-wide analysis of DNA copy-number changes using cDNA microarrays. Nat Genet. 1999, 23: 41-46. 10.1038/14385.PubMedView Article
- Shah G, Azizian M, Bruch D, Mehta R, Kittur D: Cross-species comparison of gene expression between human and porcine tissue, using single microarray platform–preliminary results. Clin Transplant. 2004, 18 (Suppl 12): 76-80. 10.1111/j.1399-0012.2004.00223.x.PubMedView Article
- von Schalburg KR, Rise ML, Cooper GA, Brown GD, Gibbs AR, Nelson CC, Davidson WS, Koop BF: Fish and chips: various methodologies demonstrate utility of a 16,006-gene salmonid microarray. BMC Genomics. 2005, 6: 126-10.1186/1471-2164-6-126.PubMedPubMed CentralView Article
- Rise ML, von Schalburg KR, Brown GD, Mawer MA, Devlin RH, Kuipers N, Busby M, Beetz-Sargent M, Alberto R, Gibbs AR, Hunt P, Shukin R, Zeznik JA, Nelson C, Jones SR, Smailus DE, Jones SJ, Schein JE, Marra MA, Butterfield YS, Stott JM, Ng SH, Davidson WS, Koop BF: Development and application of a salmonid EST database and cDNA microarray: data mining and interspecific hybridization characteristics. Genome Res. 2004, 14: 478-490. 10.1101/gr.1687304.PubMedPubMed CentralView Article
- Adjaye J, Herwig R, Herrmann D, Wruck W, Benkahla A, Brink TC, Nowak M, Carnwath JW, Hultschig C, Niemann H, Lehrach H: Cross-species hybridisation of human and bovine orthologous genes on high density cDNA microarrays. BMC Genomics. 2004, 5: 83-10.1186/1471-2164-5-83.PubMedPubMed CentralView Article
- Nieto-Diaz M, Pita-Thomas W, Nieto-Sampedro M: Cross-species analysis of gene expression in non-model mammals: reproducibility of hybridization on high density oligonucleotide microarrays. BMC Genomics. 2007, 8: 89-10.1186/1471-2164-8-89.PubMedPubMed CentralView Article
- Renn SC, Aubin-Horth N, Hofmann HA: Biologically meaningful expression profiling across species using heterologous hybridization to a cDNA microarray. BMC Genomics. 2004, 5: 42-10.1186/1471-2164-5-42.PubMedPubMed CentralView Article
- Gilad Y, Rifkin SA, Bertone P, Gerstein M, White KP: Multi-species microarrays reveal the effect of sequence divergence on gene expression profiles. Genome Res. 2005, 15: 674-680. 10.1101/gr.3335705.PubMedPubMed CentralView Article
- Bar-Or C, Bar-Eyal M, Gal TZ, Kapulnik Y, Czosnek H, Koltai H: Derivation of species-specific hybridization-like knowledge out of cross-species hybridization results. BMC Genomics. 2006, 7: 110-10.1186/1471-2164-7-110.PubMedPubMed CentralView Article
- Bar-Or C, Czosnek H, Koltai H: Cross-species microarray hybridizations: a developing tool for studying species diversity. Trends Genet. 2007, 23: 200-207. 10.1016/j.tig.2007.02.003.PubMedView Article
- Ericson PG, Johansson US: Phylogeny of Passerida (Aves: Passeriformes) based on nuclear and mitochondrial sequence data. Mol Phylogenet Evol. 2003, 29: 126-138. 10.1016/S1055-7903(03)00067-8.PubMedView Article
- Jonsson KA, Fjeldsa J, Ericson PG, Irestedt M: Systematic placement of an enigmatic Southeast Asian taxon Eupetes macrocerus and implications for the biogeography of a main songbird radiation, the Passerida. Biol Lett. 2007, 3: 323-326. 10.1098/rsbl.2007.0054.PubMedPubMed CentralView Article
- Itoh Y, Melamed E, Yang X, Kampf K, Wang S, Yehya N, Van Nas A, Replogle K, Band MR, Clayton DF, Schadt EE, Lusis AJ, Arnold AP: Dosage compensation is less effective in birds than in mammals. J Biol. 2007, 6 (1): 2-10.1186/jbiol53.PubMedPubMed CentralView Article
- Mank JE, Axelsson E, Ellegren H: Fast-X on the Z: rapid evolution of sex-linked genes in birds. Genome Res. 2007, 17: 618-624. 10.1101/gr.6031907.PubMedPubMed CentralView Article
- Huesmann G, Clayton DF: Dynamic role of postsynaptic caspase-3 and BIRC4 in zebra finch song response habituation. Neuron. 2006, 52: 1061-1072. 10.1016/j.neuron.2006.10.033.PubMedPubMed CentralView Article
- Slate J, Hale MC, Birkhead TR: Simple sequence repeats in zebra finch (Taeniopygia guttata) expressed sequence tags: a new resource for evolutionary genetic studies of passerines. BMC Genomics. 2007, 8: 52-10.1186/1471-2164-8-52.PubMedPubMed CentralView Article
- Bentley DR: Whole-genome re-sequencing. Curr Opin Genet Dev. 2006, 16: 545-552. 10.1016/j.gde.2006.10.009.PubMedView Article
- Hoheisel JD: Microarray technology: beyond transcript profiling and genotype analysis. Nat Rev Genet. 2006, 7: 200-210. 10.1038/nrg1809.PubMedView Article
- Proposal to Sequence the Genome of the Zebra Finch (Taeniopygia guttata). [http://www.genome.gov/Pages/Research/Sequencing/SeqProposals/ZebraFinchSeq2.pdf]
- Songbird Brain Transcriptome Database. [http://songbirdtranscriptome.net/]
- Wade J, Peabody C, Coussens P, Tempelman RJ, Clayton DF, Liu L, Arnold AP, Agate R: A cDNA microarray from the telencephalon of juvenile male and female zebra finches. J Neurosci Meth. 2004, 138: 199-206. 10.1016/j.jneumeth.2004.04.007.View Article
- Bonaldo MF, Lennon G, Soares MB: Normalization and subtraction: two approaches to facilitate gene discovery. Genome Res. 1996, 6: 791-806. 10.1101/gr.6.9.791.PubMedView Article
- Huang X, Madan A: CAP3: A DNA sequence assembly program. Genome Res. 1999, 9: 868-877. 10.1101/gr.9.9.868.PubMedPubMed CentralView Article
- Ensembl Chicken (Gallus gallus). [http://www.ensembl.org/Gallus_gallus/index.html]
- Gallus gallus Gene Index. [http://compbio.dfci.harvard.edu/tgi/cgi-bin/tgi/gireport.pl?gudb=g_gallus]
- Quackenbush J, Cho J, Lee D, Liang F, Holt I, Karamycheva S, Parvizi B, Pertea G, Sultana R, White J: The TIGR Gene Indices: analysis of gene transcript sequences in highly sampled eukaryotic species. Nucleic Acids Res. 2001, 29: 159-164. 10.1093/nar/29.1.159.PubMedPubMed CentralView Article
- Stripling R, Milewski L, Kruse AA, Clayton DF: Rapidly learned song-discrimination without behavioral reinforcement in adult male zebra finches (Taeniopygia guttata). Neurobiol Learn Mem. 2003, 79: 41-50. 10.1016/S1074-7427(02)00005-9.PubMedView Article
- Cheng HY, Clayton DF: Activation and habituation of extracellular signal-regulated kinase phosphorylation in zebra finch auditory forebrain during song presentation. J Neurosci. 2004, 24: 7503-7513. 10.1523/JNEUROSCI.1405-04.2004.PubMedView Article
- Cui X, Churchill GA: Statistical tests for differential expression in cDNA microarray experiments. Genome Biol. 2003, 4: 210-10.1186/gb-2003-4-4-210.PubMedPubMed CentralView Article
- Kerr MK, Martin M, Churchill GA: Analysis of variance for gene expression microarray data. J Comput Biol. 2000, 7: 819-837. 10.1089/10665270050514954.PubMedView Article
- Wolfinger RD, Gibson G, Wolfinger ED, Bennett L, Hamadeh H, Bushel P, Afshari C, Paules RS: Assessing gene significance from cDNA microarray expression data via mixed models. J Comput Biol. 2001, 8: 625-637. 10.1089/106652701753307520.PubMedView Article
- Rodriguez-Zas SL: Comparison of statistical methods to study cDNA microarray data. J Dairy Sci. 2002, 80: 10-
- Rodriguez-Zas SL, Southey BR: Linear mixed effects models for microarray gene expression data. Proceedings of the World Congress in Genetics Applied to Livestock Production. 2002
- Bioconductor. [http://www.bioconductor.org]
- SCIBLU, LTH: AFFYMETRIX MICROARRAYS. [http://www.lth.se/index.php?id=7789]
- Smyth GK: Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol. 2004, 3: Article3-PubMed
- Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP: Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res. 2003, 31: e15-10.1093/nar/gng015.PubMedPubMed CentralView Article
- Benjamini YYH: Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Royal Statistical Soc Series B (Methodological). 1995, 57: 289-300.
- Tusher VG, Tibshirani R, Chu G: Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001, 98: 5116-5121. 10.1073/pnas.091062498.PubMedPubMed CentralView Article
- CUGI. [http://www.genome.clemson.edu]
- ARDAS. [http://126.96.36.199:8080/ardas/install.html]
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.