Panel sequencing of 264 candidate susceptibility genes and segregation analysis in a cohort of non-BRCA1, non-BRCA2 breast cancer families.

PURPOSE: The main aim of this study was to screen epigenetic modifier genes and known breast cancer driver genes for germline mutations in non-BRCA1/2 (BRCAx) breast cancer families in order to identify novel susceptibility genes of moderate-high penetrance. METHODS: We screened 264 candidate susceptibility genes in 656 index cases from non-BRCA1/2 families. Potentially pathogenic candidate mutations were then genotyped in all available family members for the assessment of co-segregation of the variant with disease in the family in order to estimate the breast cancer risks associated with these mutations. For 11 of the candidate susceptibility genes, we screened an additional 800 non-BRCA1/2 breast cancer cases and 787 controls. RESULTS: Only two genes, CHD8 and USH2A showed any evidence of an increased risk of breast cancer (RR = 2.40 (95% CI 1.0-7.32) and 2.48 (95% CI 1.11-6.67), respectively). CONCLUSIONS: We found no convincing evidence that epigenetic modifier and known breast cancer driver genes carry germline mutations that increase breast cancer risk. USH2A is no longer regarded as a breast cancer driver gene and seems an implausible candidate given its association with Usher syndrome. However, somatic mutations in CHD8 have been recently reported, making it an even more promising candidate, but further analysis of CHD8 in very large cohorts of families or case-control studies would be required to determine if it is a moderate-risk breast cancer susceptibility gene.

Targeted massively parallel sequencing of a panel of putative breast cancer susceptibility genes in a large cohort of multiple-case breast and ovarian cancer families.

INTRODUCTION: Gene panel testing for breast cancer susceptibility has become relatively cheap and accessible. However, the breast cancer risks associated with mutations in many genes included in these panels are unknown. METHODS: We performed custom-designed targeted sequencing covering the coding exons of 17 known and putative breast cancer susceptibility genes in 660 non-BRCA1/2 women with familial breast cancer. Putative deleterious mutations were genotyped in relevant family members to assess co-segregation of each variant with disease. We used maximum likelihood models to estimate the breast cancer risks associated with mutations in each of the genes. RESULTS: We found 31 putative deleterious mutations in 7 known breast cancer susceptibility genes (TP53, PALB2, ATM, CHEK2, CDH1, PTEN and STK11) in 45 cases, and 22 potential deleterious mutations in 31 cases in 8 other genes (BARD1, BRIP1, MRE11, NBN, RAD50, RAD51C, RAD51D and CDK4). The relevant variants were then genotyped in 558 family members. Assuming a constant relative risk of breast cancer across age groups, only variants in CDH1, CHEK2, PALB2 and TP53 showed evidence of a significantly increased risk of breast cancer, with some supportive evidence that mutations in ATM confer moderate risk. CONCLUSIONS: Panel testing for these breast cancer families provided additional relevant clinical information for <2% of families. We demonstrated that segregation analysis has some potential to help estimate the breast cancer risks associated with mutations in breast cancer susceptibility genes, but very large case-control sequencing studies and/or larger family-based studies will be needed to define the risks more accurately.

An Ariadne's thread to the identification and annotation of noncoding RNAs in eukaryotes.

Non-protein coding RNAs (ncRNAs) have emerged as a vast and heterogeneous portion of eukaryotic transcriptomes. Several ncRNA families, either short (<200 nucleotides, nt) or long (>200 nt), have been described and implicated in a variety of biological processes, from translation to gene expression regulation and nuclear trafficking. Most probably, other families are still to be discovered. Computational methods for ncRNA research require different approaches from the ones normally used in the prediction of protein-coding genes. Indeed, primary sequence alone is often insufficient to infer ncRNA functionality, whereas secondary structure and local conservation of portions of the transcript could provide useful information for both the prediction and the functional annotation of ncRNAs. Here we present an overview of computational methods and bioinformatics resources currently available for studying ncRNA genes, introducing the common themes as well as the different approaches required for long and short ncRNA identification and annotation.

Identification of conserved Drosophila-specific euchromatin-restricted non-coding sequence motifs.

Non-protein-coding DNA comprises the majority of animal genomes but its functions are largely unknown. We identified over 17,000 different tetranucleotide pairs in the Drosophila melanogaster genome that are over-represented at distances up to 100nt in conserved non-exonic sequences. Those exhibiting the highest information content in surrounding nucleotides were classified into five groups: tRNAs, motifs associated with histone genes, Suppressor-of-Hairy-wing binding sites, and two sets of previously unrecognized motifs (DLM3 and DLM4). There are hundreds to thousands of copies of DLM3 and DLM4, respectively, in the genome, located almost exclusively in non-coding regions. They have similar copy numbers among drosophilids, but are largely absent in other insects. DLM3 is likely a cis-regulatory element, whereas DLM4 sequences are capable of forming a short hairpin structure and are expressed as approximately 80nt RNAs. This work reports the existence of Drosophila genus-specific sequence motifs, and suggests that many more novel functional elements may be discovered in genomes using the general approach outlined herein.

Identification of novel non-coding RNAs using profiles of short sequence reads from next generation sequencing data.

BACKGROUND: The increasing interest in small non-coding RNAs (ncRNAs) such as microRNAs (miRNAs), small interfering RNAs (siRNAs) and Piwi-interacting RNAs (piRNAs) and recent advances in sequencing technology have yielded large numbers of short (18-32 nt) RNA sequences from different organisms, some of which are derived from small nucleolar RNAs (snoRNAs) and transfer RNAs (tRNAs). We observed that these short ncRNAs frequently cover the entire length of annotated snoRNAs or tRNAs, which suggests that other loci specifying similar ncRNAs can be identified by clusters of short RNA sequences. RESULTS: We combined publicly available datasets of tens of millions of short RNA sequence tags from Drosophila melanogaster, and mapped them to the Drosophila genome. Approximately 6 million perfectly mapping sequence tags were then assembled into 521,302 tag-contigs (TCs) based on tag overlap. Most transposon-derived sequences, exons and annotated miRNAs, tRNAs and snoRNAs are detected by TCs, which show distinct patterns of length and tag-depth for different categories. The typical length and tag-depth of snoRNA-derived TCs was used to predict 7 previously unrecognized box H/ACA and 26 box C/D snoRNA candidates. We also identified one snRNA candidate and 86 loci with a high number of tags that are yet to be annotated, 7 of which have a particular 18mer motif and are located in introns of genes involved in development. A subset of new snoRNA candidates and putative ncRNA candidates was verified by Northern blot. CONCLUSIONS: In this study, we have introduced a new approach to identify new members of known classes of ncRNAs based on the features of TCs corresponding to known ncRNAs. A large number of the identified TCs are yet to be examined experimentally suggesting that many more novel ncRNAs remain to be discovered.

Paucity and preferential suppression of transgenes in late replication domains of the D. melanogaster genome.

BACKGROUND: Eukaryotic genomes are organized in extended domains with distinct features intimately linking genome structure, replication pattern and chromatin state. Recently we identified a set of long late replicating euchromatic regions that are underreplicated in salivary gland polytene chromosomes of D. melanogaster. RESULTS: Here we demonstrate that these underreplicated regions (URs) have a low density of P-element and piggyBac insertions compared to the genome average or neighboring regions. In contrast, Minos-based transposons show no paucity in URs but have a strong bias to testis-specific genes. We estimated the suppression level in 2,852 stocks carrying a single P-element by analysis of eye color determined by the mini-white marker gene and demonstrate that the proportion of suppressed transgenes in URs is more than three times higher than in the flanking regions or the genomic average. The suppressed transgenes reside in intergenic, genic or promoter regions of the annotated genes. We speculate that the low insertion frequency of P-elements and piggyBacs in URs partially results from suppression of transgenes that potentially could prevent identification of transgenes due to complete suppression of the marker gene. In a similar manner, the proportion of suppressed transgenes is higher in loci replicating late or very late in Kc cells and these loci have a lower density of P-elements and piggyBac insertions. In transgenes with two marker genes suppression of mini-white gene in eye coincides with suppression of yellow gene in bristles. CONCLUSIONS: Our results suggest that the late replication domains have a high inactivation potential apparently linked to the silenced or closed chromatin state in these regions, and that such inactivation potential is largely maintained in different tissues.

Large-scale appearance of ultraconserved elements in tetrapod genomes and slowdown of the molecular clock.

Mammalian genomes contain millions of highly conserved noncoding sequences, many of which are regulatory. The most extreme examples are the 481 ultraconserved elements (UCEs) that are identical over at least 200 bp in human, mouse, and rat and show 96% identity with chicken, which diverged approximately 310 MYA. If the substitution rate in UCEs remained constant, these elements should also be present with a high level of identity in fish (approximately 450 Myr), but this is not the case, suggesting that many appeared in the amniotes or tetrapods or that the molecular clock has slowed down in these lineages, or both. Taking advantage of the availability of multiple genomes, we identified 13,736 UCEs in the human genome that are identical over at least 100 bp in at least 3 of 5 placental mammals, including 2,189 sequences over at least 200 bp, thereby greatly expanding the repertoire of known UCEs, and investigated the evolution of these sequences in opossum, chicken, frog, and fish. We conclude that there was a massive genome-wide acquisition and expansion of UCEs during tetrapod and then amniote evolution, accompanied by a slowdown of the molecular clock, particularly in the amniotes, a process consistent with their functional exaptation in these lineages. The majority of tetrapod-specific UCEs are noncoding and associated with genes involved in regulation of transcription and development. In contrast, fish genomes contain relatively few UCEs, the majority of which are common to all bony vertebrates. These elements are different from other conserved noncoding elements and appear to be important regulatory innovations that became fixed following the emergence of vertebrates from the sea to the land.

Maintenance of transposon-free regions throughout vertebrate evolution.

BACKGROUND: We recently reported the existence of large numbers of regions up to 80 kb long that lack transposon insertions in the human, mouse and opossum genomes. These regions are significantly associated with loci involved in developmental and transcriptional regulation. RESULTS: Here we report that transposon-free regions (TFRs) are prominent genomic features of amphibian and fish lineages, and that many have been maintained throughout vertebrate evolution, although most transposon-derived sequences have entered these lineages after their divergence. The zebrafish genome contains 470 TFRs over 10 kb and a further 3,951 TFRs over 5 kb, which is comparable to the number identified in mammals. Two thirds of zebrafish TFRs over 10 kb are orthologous to TFRs in at least one mammal, and many have orthologous TFRs in all three mammalian genomes as well as in the genome of Xenopus tropicalis. This indicates that the mechanism responsible for the maintenance of TFRs has been active at these loci for over 450 million years. However, the majority of TFR bases cannot be aligned between distantly related species, demonstrating that TFRs are not the by-product of strong primary sequence conservation. Syntenically conserved TFRs are also more enriched for regulatory genes compared to lineage-specific TFRs. CONCLUSION: We suggest that TFRs contain extended regulatory sequences that contribute to the precise expression of genes central to early vertebrate development, and can be used as predictors of important regulatory regions.

Genomics Virtual Laboratory: A Practical Bioinformatics Workbench for the Cloud.

BACKGROUND: Analyzing high throughput genomics data is a complex and compute intensive task, generally requiring numerous software tools and large reference data sets, tied together in successive stages of data transformation and visualisation. A computational platform enabling best practice genomics analysis ideally meets a number of requirements, including: a wide range of analysis and visualisation tools, closely linked to large user and reference data sets; workflow platform(s) enabling accessible, reproducible, portable analyses, through a flexible set of interfaces; highly available, scalable computational resources; and flexibility and versatility in the use of these resources to meet demands and expertise of a variety of users. Access to an appropriate computational platform can be a significant barrier to researchers, as establishing such a platform requires a large upfront investment in hardware, experience, and expertise. RESULTS: We designed and implemented the Genomics Virtual Laboratory (GVL) as a middleware layer of machine images, cloud management tools, and online services that enable researchers to build arbitrarily sized compute clusters on demand, pre-populated with fully configured bioinformatics tools, reference datasets and workflow and visualisation options. The platform is flexible in that users can conduct analyses through web-based (Galaxy, RStudio, IPython Notebook) or command-line interfaces, and add/remove compute nodes and data resources as required. Best-practice tutorials and protocols provide a path from introductory training to practice. The GVL is available on the OpenStack-based Australian Research Cloud (http://nectar.org.au) and the Amazon Web Services cloud. The principles, implementation and build process are designed to be cloud-agnostic. CONCLUSIONS: This paper provides a blueprint for the design and implementation of a cloud-based Genomics Virtual Laboratory. We discuss scope, design considerations and technical and logistical constraints, and explore the value added to the research community through the suite of services and resources provided by our implementation.

Underreplicated regions in Drosophila melanogaster are enriched with fast-evolving genes and highly conserved noncoding sequences.

Many late replicating regions are underreplicated in polytene chromosomes of Drosophila melanogaster. These regions contain silenced chromatin and overlap long syntenic blocks of conserved gene order in drosophilids. In this report we show that in D. melanogaster the underreplicated regions are enriched with fast-evolving genes lacking homologs in distant species such as mosquito or human, indicating that the phylogenetic conservation of genes correlates with replication timing and chromatin status. Drosophila genes without human homologs located in the underreplicated regions have higher nonsynonymous substitution rate and tend to encode shorter proteins when compared with those in the adjacent regions. At the same time, the underreplicated regions are enriched with ultraconserved elements and highly conserved noncoding sequences, especially in introns of very long genes indicating the presence of an extensive regulatory network that may be responsible for the conservation of gene order in these regions. The regions have a modest preference for long noncoding RNAs but are depleted for small nucleolar RNAs, microRNAs, and transfer RNAs. Our results demonstrate that the underreplicated regions have a specific genic composition and distinct pattern of evolution.

Rare mutations in RINT1 predispose carriers to breast and Lynch syndrome-spectrum cancers.

UNLABELLED: Approximately half of the familial aggregation of breast cancer remains unexplained. A multiple-case breast cancer family exome-sequencing study identified three likely pathogenic mutations in RINT1 (NM_021930.4) not present in public sequencing databases: RINT1 c.343C>T (p.Q115X), c.1132_1134del (p.M378del), and c.1207G>T (p.D403Y). On the basis of this finding, a population-based case-control mutation-screening study was conducted that identified 29 carriers of rare (minor allele frequency < 0.5%), likely pathogenic variants: 23 in 1,313 early-onset breast cancer cases and six in 1,123 frequency-matched controls [OR, 3.24; 95% confidence interval (CI), 1.29-8.17; P = 0.013]. RINT1 mutation screening of probands from 798 multiple-case breast cancer families identified four additional carriers of rare genetic variants. Analysis of the incidence of first primary cancers in families of women carrying RINT1 mutations estimated that carriers were at increased risk of Lynch syndrome-spectrum cancers [standardized incidence ratio (SIR), 3.35; 95% CI, 1.7-6.0; P = 0.005], particularly for relatives diagnosed with cancer under the age of 60 years (SIR, 10.9; 95% CI, 4.7-21; P = 0.0003). SIGNIFICANCE: The work described in this study adds RINT1 to the growing list of genes in which rare sequence variants are associated with intermediate levels of breast cancer risk. Given that RINT1 is also associated with a spectrum of cancers with mismatch repair defects, these findings have clinical applications and raise interesting biological questions.

Comparison of ultra-conserved elements in drosophilids and vertebrates.

Metazoan genomes contain many ultra-conserved elements (UCEs), long sequences identical between distant species. In this study we identified UCEs in drosophilid and vertebrate species with a similar level of phylogenetic divergence measured at protein-coding regions, and demonstrated that both the length and number of UCEs are larger in vertebrates. The proportion of non-exonic UCEs declines in distant drosophilids whilst an opposite trend was observed in vertebrates. We generated a set of 2,126 Sophophora UCEs by merging elements identified in several drosophila species and compared these to the eutherian UCEs identified in placental mammals. In contrast to vertebrates, the Sophophora UCEs are depleted around transcription start sites. Analysis of 52,954 P-element, piggyBac and Minos insertions in the D. melanogaster genome revealed depletion of the P-element and piggyBac insertions in and around the Sophophora UCEs. We examined eleven fly strains with transposon insertions into the intergenic UCEs and identified associated phenotypes in five strains. Four insertions behave as recessive lethals, and in one case we observed a suppression of the marker gene within the transgene, presumably by silenced chromatin around the integration site. To confirm the lethality is caused by integration of transposons we performed a phenotype rescue experiment for two stocks and demonstrated that the excision of the transposons from the intergenic UCEs restores viability. Sequencing of DNA after the transposon excision in one fly strain with the restored viability revealed a 47 bp insertion at the original transposon integration site suggesting that the nature of the mutation is important for the appearance of the phenotype. Our results suggest that the UCEs in flies and vertebrates have both common and distinct features, and demonstrate that a significant proportion of intergenic drosophila UCEs are sensitive to disruption.

Late replication domains are evolutionary conserved in the Drosophila genome.

Drosophila chromosomes are organized into distinct domains differing in their predominant chromatin composition, replication timing and evolutionary conservation. We show on a genome-wide level that genes whose order has remained unaltered across 9 Drosophila species display late replication timing and frequently map to the regions of repressive chromatin. This observation is consistent with the existence of extensive domains of repressive chromatin that replicate extremely late and have conserved gene order in the Drosophila genome. We suggest that such repressive chromatin domains correspond to a handful of regions that complete replication at the very end of S phase. We further demonstrate that the order of genes in these regions is rarely altered in evolution. Substantial proportion of such regions significantly coincide with large synteny blocks. This indicates that there are evolutionary mechanisms maintaining the integrity of these late-replicating chromatin domains. The synteny blocks corresponding to the extremely late-replicating regions in the D. melanogaster genome consistently display two-fold lower gene density across different Drosophila species.

Conservation of domain structure in a fast-evolving heterochromatic SUUR protein in drosophilids.

Different genomic regions replicate at a distinct time during S-phase. The SuUR mutation alters replication timing and the polytenization level of intercalary and pericentric heterochromatin in Drosophila melanogaster salivary gland polytene chromosomes. We analyzed SuUR in different insects, identified conserved regions in the protein, substituted conserved amino acid residues, and studied effects of the mutations on SUUR function. SuUR orthologs were identified in all sequenced drosophilids, and a highly divergent ortholog was found in the mosquito genome. We demonstrated that SUUR evolves at very high rate comparable with that of Transformer. Remarkably, upstream ORF within 5' UTR of the gene is more conserved than SUUR in drosophilids, but it is absent in the mosquito. The domain structure and charge of SUUR are maintained in drosophilids despite the high divergence of the proteins. The N-terminal part of SUUR with similarity to the SNF2/SWI2 proteins displays the highest level of conservation. Mutation of two conserved amino acid residues in this region impairs binding of SUUR to polytene chromosomes and reduces the ability of the protein to cause DNA underreplication. The least conserved middle part of SUUR interacting with HP1 retains positively and negatively charged clusters and nuclear localization signals. The C terminus contains interlacing conserved and variable motifs. Our results suggest that SUUR domains evolve with different rates and patterns but maintain their features.

Orthologous microRNA genes are located in cancer-associated genomic regions in human and mouse.

BACKGROUND: MicroRNAs (miRNAs) are short non-coding RNAs that regulate differentiation and development in many organisms and play an important role in cancer. METHODOLOGY/PRINCIPAL FINDINGS: Using a public database of mapped retroviral insertion sites from various mouse models of cancer we demonstrate that MLV-derived retroviral inserts are enriched in close proximity to mouse miRNA loci. Clustered inserts from cancer-associated regions (Common Integration Sites, CIS) have a higher association with miRNAs than non-clustered inserts. Ten CIS-associated miRNA loci containing 22 miRNAs are located within 10 kb of known CIS insertions. Only one CIS-associated miRNA locus overlaps a RefSeq protein-coding gene and six loci are located more than 10 kb from any RefSeq gene. CIS-associated miRNAs on average are more conserved in vertebrates than miRNAs associated with non-CIS inserts and their human homologs are also located in regions perturbed in cancer. In addition we show that miRNA genes are enriched around promoter and/or terminator regions of RefSeq genes in both mouse and human. CONCLUSIONS/SIGNIFICANCE: We provide a list of ten miRNA loci potentially involved in the development of blood cancer or brain tumors. There is independent experimental support from other studies for the involvement of miRNAs from at least three CIS-associated miRNA loci in cancer development.

Non-coding RNA.

The term non-coding RNA (ncRNA) is commonly employed for RNA that does not encode a protein, but this does not mean that such RNAs do not contain information nor have function. Although it has been generally assumed that most genetic information is transacted by proteins, recent evidence suggests that the majority of the genomes of mammals and other complex organisms is in fact transcribed into ncRNAs, many of which are alternatively spliced and/or processed into smaller products. These ncRNAs include microRNAs and snoRNAs (many if not most of which remain to be identified), as well as likely other classes of yet-to-be-discovered small regulatory RNAs, and tens of thousands of longer transcripts (including complex patterns of interlacing and overlapping sense and antisense transcripts), most of whose functions are unknown. These RNAs (including those derived from introns) appear to comprise a hidden layer of internal signals that control various levels of gene expression in physiology and development, including chromatin architecture/epigenetic memory, transcription, RNA splicing, editing, translation and turnover. RNA regulatory networks may determine most of our complex characteristics, play a significant role in disease and constitute an unexplored world of genetic variation both within and between species.

Transposon-free regions in mammalian genomes.

Despite the presence of over 3 million transposons separated on average by approximately 500 bp, the human and mouse genomes each contain almost 1000 transposon-free regions (TFRs) over 10 kb in length. The majority of human TFRs correlate with orthologous TFRs in the mouse, despite the fact that most transposons are lineage specific. Many human TFRs also overlap with orthologous TFRs in the marsupial opossum, indicating that these regions have remained refractory to transposon insertion for long evolutionary periods. Over 90% of the bases covered by TFRs are noncoding, much of which is not highly conserved. Most TFRs are not associated with unusual nucleotide composition, but are significantly associated with genes encoding developmental regulators, suggesting that they represent extended regions of regulatory information that are largely unable to tolerate insertions, a conclusion difficult to reconcile with current conceptions of gene regulation.

Small regulatory RNAs in mammals.

Mammalian cells harbor numerous small non-protein-coding RNAs, including small nucleolar RNAs (snoRNAs), microRNAs (miRNAs), short interfering RNAs (siRNAs) and small double-stranded RNAs, which regulate gene expression at many levels including chromatin architecture, RNA editing, RNA stability, translation, and quite possibly transcription and splicing. These RNAs are processed by multistep pathways from the introns and exons of longer primary transcripts, including protein-coding transcripts. Most show distinctive temporal- and tissue-specific expression patterns in different tissues, including embryonal stem cells and the brain, and some are imprinted. Small RNAs control a wide range of developmental and physiological pathways in animals, including hematopoietic differentiation, adipocyte differentiation and insulin secretion in mammals, and have been shown to be perturbed in cancer and other diseases. The extent of transcription of non-coding sequences and the abundance of small RNAs suggests the existence of an extensive regulatory network on the basis of RNA signaling which may underpin the development and much of the phenotypic variation in mammals and other complex organisms and which may have different genetic signatures from sequences encoding proteins.

Genomic analysis of Drosophila chromosome underreplication reveals a link between replication control and transcriptional territories.

In Drosophila polytene chromosomes, most late-replicating regions remain underreplicated. A loss-of-function mutant of the suppressor of underreplication [Su(UR)] gene suppresses underreplication (UR), whereas extra copies of this gene enhance the level and number of regions showing UR. By combining DNA microarray analysis with manipulation of the number of Su(UR) gene copies, we achieved genomic-scale molecular identification of 1,036 genes that are arranged in clusters located in 52 UR chromosomal regions. These regions overlap extensively (96%) but are not completely identical with late-replicating regions of mitotically dividing Kc cells in culture. Reanalysis of published gene expression profiles revealed that genomic regions defined by replication properties include clusters of coordinately expressed genes. Genomic regions that are UR in polytene chromosomes and late replicated in Kc cell chromosomes show a particularly common association with transcriptional territories that are expressed in testis/males but not ovary/females or embryos. An attractive hypothesis for future testing is that factors involved in replication control, such as SU(UR), may interact physically with those involved in epigenetic silencing of transcription territories.

Ultraconserved elements in the human genome.

There are 481 segments longer than 200 base pairs (bp) that are absolutely conserved (100% identity with no insertions or deletions) between orthologous regions of the human, rat, and mouse genomes. Nearly all of these segments are also conserved in the chicken and dog genomes, with an average of 95 and 99% identity, respectively. Many are also significantly conserved in fish. These ultraconserved elements of the human genome are most often located either overlapping exons in genes involved in RNA processing or in introns or nearby genes involved in the regulation of transcription and development. Along with more than 5000 sequences of over 100 bp that are absolutely conserved among the three sequenced mammals, these represent a class of genetic elements whose functions and evolutionary origins are yet to be determined, but which are more highly conserved between these species than are proteins and appear to be essential for the ontogeny of mammals and other vertebrates.