Field of the invention

The present invention relates to aquaculture, and particularly to aquaculture of fish, smolt production of salmonid fish included. Further, the invention relates to methods to identify fish characteristics and for providing robust and high-quality farmed fish.

Background of the invention

Although fish farming, and particularly the aquaculture of salmonid species, is considered as a successful young industry in many temperate countries, with Norway as a leading player, there are still many challenges. Several of these challenges are related to fish health and welfare and to environmental issues, and the challenges have considerable impact on the economics and the sustainability of the sector.

Major components of the problem are infectious diseases, parasite infestations like sea lice, developmental malformations, highly variable growth rates in the sea phase, and adverse environmental impacts including fish escapes, wastes, transfer of parasites and infections to wild stocks.

Further, salmon industry is still immature in the sense that most of the marketing volume is in the form of bulk products and less as value-added and branded seafood. This means that this quite volatile/price sensitive sector with very delicate production regimes in the pre-harvest part is not taking full advantage of the margin opportunities which could have been enjoyed with branded higher priced products. Moreover, the existing brands are so far not protected sufficiently by biological markers verifying their origin.

Presently, the above-mentioned problems have been sought to be solved through multiple sets of preventive and curative regimes. These may consist of rules and regulations from authorities linked with permits/licenses (quarantines, max admitted biomass, locations), fish welfare and best practice managements, breeding, vaccines, feeds and feed ingredients, treatments, optimized gears and infrastructure, optimized managements. Despite of all these efforts, the fish farming sector is still suffering from the described challenges, some of which even have escalated over the recent years.

One part of the above-described complex problem is related to inferior seed quality, i.e. smolt in salmon farming. The smolt is often not as robust as desired, mostly due to suboptimal smolt production regimes. Of the two major smolt production regimes Flowthrough (FT) systems and Recirculation aquaculture systems (RAS), RAS still experience more challenges than FT, although RAS are regarded as more sustainable. Even though RAS are superior to FT as of environmental footprint and have future big potentials in fish farming, its complexity imposes continuous improvements on e.g. water treatment without adverse effects on the fish as a main challenge. The sector is still experiencing superior sea phase performance of FT raised smolt compared to RAS raised. Methods to optimise production protocols (rearing protocols) of smolt production regimes are needed.

Moreover, the existing test systems aimed at reflecting robustness, maturation, life phase readiness etc. in fish farming, such as in the smolt production, are inferior. For instance, the current methods to evaluate the entire smolt production regimes, critical phases included like the smoltification process (a metamorphic process) and smolt window timing, are not robust enough simply due to the fact that they are restricted to the testing of single or too few biological markers compared to the complexity of the mentioned biological processes. The nature of the fish and smolt quality is resulting from a vast number of systems’ biology interplaying factors and pathways under varying genetic control and environmental influence, the latter also embracing production regimes with accompanying protocols. As for the smoltification part; not only are a lot of bio-elements and biorhythms involved but many of them also need to be synchronized. Flence, there is a need for an analytical method that can reflect the fish maturation status and robustness in its various life phases in a better way, and also which in turn is able to provide feedback to optimizing the fish production and accompanying management regimes in such. Brief summary of the invention

The invention provides methods for farmed fish handling or production comprising the provision and analysis of epigenetic signatures of fish sample materials. The invention provides methods to identify farmed fish characteristics or its origin, by the provision and analysis of epigenetic signatures of fish sample material. The epigenetic signatures are prepared from selected groups of genes or specific genes.

A key element of the invention is the uncovering of highly contrasting epigenetic signatures of genes of farmed fish when comparing the two major smolt production regimes flow through (FT) and recirculation aquaculture systems (RAS). The contrasts in the signatures may serve as a novel objective guide to optimise production protocols of regimes.

Another important part of the invention is the discovery and description of novel gene specific epigenetic contrasts between early development phases of salmon. This biological development clock, “SalmoClock”, may serve at optimizing the critical early phases of the cultivation.

Hence, by the provision and analysis of epigenetic signatures of fish sample materials one can identify different fish production regimes with their different fish characteristics or their origin, location included, since location will be interlinked both with production regime and with unique environmental characteristics with impact on the epigenome.

In one aspect the invention provides a method to identify fish characteristics of farmed fish, comprising a step of preparing at least one epigenetic signature from a fish sample material, wherein the method comprises the steps of: i) sampling to obtain fish sample material; ii) DNA sequencing, comprising carrying out genome sequencing of the fish sample material; iii) analysing the revealed genome data set of step ii) and establishing epigenetic signatures for the samples; and optionally iv) comparing and correlating the epigenetic signatures obtained with existing epigenetic signatures; wherein the epigenetic signatures are prepared from either of the genes disclosed herein, such as from either of those included in Example 1 , Tables 1a, 1b, 2a-d.

The prepared epigenetic signatures of the fish sample material may be for use as authenticators for fish. Such use may include traceability of fish, for use as a verification of a given fish production protocol/regime, such as for use in determination of the origin of escaped farmed fish, and for optimising production protocols of regimes. Further, the prepared epigenetic signatures may form part of an epigenetic signature-based test and enhancement system for either of fish welfare and qualities; robustness, maturation, biological age, authentification, in vaccinology and other disease preventive measures or in breeding regimes.

Brief description of the drawings:

Figure 1 provides a violin plot demonstrating the distribution of methylation values for genes/features of salmon, reference is made to the study of Example 1.

Figure 2 illustrates the “SalmoClock”concept, providing an early phase biological/development methyl clock. This employ gene markers displaying increasing as well as decreasing methylation levels, on the y-axis (0-1.0) during 5 development phases, on the x-axis (0= fertilized eggs, 1= yolk sack larva, 2= fry, 3= parr and 4= smolt) and over 4 comparison steps between these phases. Adequate statistics and algorithms like e.g. regression analysis (like e.g. elastic net) and machine learning have been employed to continuously select informative markers/genes and to optimize a composite regression and correlation between phases and methylation levels.

Detailed description of the invention

The method of the invention comprises the provision and analysis of epigenetic signatures of fish sample materials, particularly based on certain groups of genes or from specific genes. The inventors have discovered that the provision and analysis of epigenetic signatures in fish farming, as disclosed in PCT/N02021/050030 of the applicant, is a concept that may be very useful as further presented herein. Whereas PCT/N02021/050030 provides a broad description of the concept of providing and using epigenetic signatures in fish farming, the inventors have now surprisingly found that certain groups of genes, or even specific genes, should be selected as basis for the epigenetic signatures.

In one aspect the invention provides a method to identify fish characteristics of farmed fish, comprising a step of preparing at least one epigenetic signature from a fish sample material.

In one embodiment, the invention takes advantage of the finding that there are contrasting epigenetic signatures for certain genes of fish from RAS and FT production regimes. The inventors have uncovered highly contrasting epigenetic signatures, i.e. methylation levels of genes, when comparing the two smolt production regimes Flowthrough (FT) and Recirculation aquaculture systems (RAS). The contrasts in the signatures may for example serve as a novel objective guide to optimise production protocols of regimes. For instance, the difference in epigenetic signatures from certain genes may be useful in improving the RAS systems based on mentioned contrasts since this regime is less mature and less well proven than FT and is phasing more challenges in the grow out phase than the FT system. The FT system may hence be regarded as a gold standard closer to the natural conditions of salmonid environments upon which other less mature industrial systems can take advantage of through epigenetic guidance.

In another embodiment, the invention takes advantage of the finding that there are contrasting epigenetic signatures for certain genes from one salmon development phase to another. The inventors have uncovered novel gene specific epigenetic contrasts between early development phases of salmon, i.e. from egg to smolt.

These epigenetic modifications taking place from one life phase to another, as a biological development clock, called the “SalmoClock” herein, may serve at optimizing the critical early phases of the cultivation of salmon where major premises is laid down for a successful grow out phase.

Two major smolt production regimes are predominant, namely Flowthrough (FT) and Recirculation aquaculture systems (RAS). Although there is a series of variables or parameters that change within the two systems, it is expected that when comparing epigenetic markers (signatures) between these two regimes, one will find discrepancies in such markers of outmost industrial relevance. This is true both because of the existence of considerable difference between the regimes in terms of many parameters but also because of the experience from the aquaculture sector that RAS is suffering from more challenges than the other, and finally, that any substantial further expansion of aquaculture needs to come from RAS due to sustainability requirements, water restrictions, as well as lack of access to FT locations.

Flence, the inventors focused its strategy on uncovering epigenetic informative differences between FT and RAS systems and to reveal epigenetic informative markers of and between critical early life phases with emphasis on methylome analysis.

Epigenetics refers to the study of heritable changes that do not involve changes to the DNA sequence, but which have resulted from chemical modifications to the DNA, chromatin remodeling, histone modification or noncoding RNA which affect gene expression. Epigenetic marks, or signatures, which act on expression both in time (“biological clock”) and space (tissue development), can be inherited to progeny of cells (epigenetic memory) or to progeny of organisms (transgenerational inheritance). In the former case it is a major driver of cell differentiation and the development of an individual’s entire life span. In the latter case the marks are imprinted in the germline genome (sperm or egg cells) of the parents and may bestow parent specific “messages” to govern the expression of selected genes of the offspring provided the marks overcome the gametogenic and embryonic reprogramming which normally takes place at a high degree. Part of the marks do overcome this reprogramming, which is the reason why epigenomes can display transgenerational inheritance. A considerable number of genes of vertebrates are differentially expressed in the offspring related to the parent of origin: a copy (allele) of a specific gene inherited from one parent may be expressed whereas the other allele of the same gene inherited from the other parent may be non-expressed. This parent of origin specific expression is called genomic imprinting. Appropriate imprinting of specific genes is important for normal development.

Epigenetic marks are intimately linked with the biorhythms of an individual and can change, i.e. become reprogrammed, in response to environmental stimuli over the course of an organism's life. The applicant has now found that details of the management regimes or environmental effectors during the pre-harvest fish farming phase affect the epigenetics and render epigenetic signatures in the organism, and this fact may be used in farmed fish handling or production, smolt production included.

One type of epigenetic mechanism is DNA methylation, where a methyl group (Chte-) is added to bases of genomic DNA by specific enzymes, DNA-methyltransferases. Two of DNA's four bases, cytosine and adenine, can be methylated. Usually it is the base Cytosine (C) which is methylated at the same 5-position of the pyrimidine ring (5mC) or hydroxymethylated (5hmC), and most often the cytosine residue is followed by a guanine residue, forming a CpG site. Sequences enriched in CpG sites, called CpG islands, often surround promoters and are typical sites for transcription initiation. Methylation can change gene expression. When located in a gene promoter, DNA methylation typically acts to repress gene transcription. When located in the gene body, methylation may enhance gene expression, especially if the CpG islands in the promoter region are not methylated or hypomethylated. DNA methylation is essential for normal development and is associated with a number of key processes including genomic imprinting, X-chromosome inactivation, repression of transposable elements, aging, carcinogenesis, and cell differentiation.

At least for humans and other mammals, it is known that DNA methylation levels can be used to estimate the chronological age of tissues, cell types and individuals based on their biological age. The latter is obtained by correlating the shifting methylation pattern with time and cell divisions, and hence forming an accurate epigenetic clock.

Definitions:

By the term “methylome”, we mean the set of methylated modifications of the DNA of an organism’s genome or cells. Methylation is a major chemical change within the term epigenome (see below) to change the function of a genome.

By the term “epigenome”, we mean chemical changes to the DNA and histone proteins of an organism that can be passed down to progeny of cells and organisms. Such changes can lead to functional changes of the genome like influencing gene expression, while the term “transcriptome” means the set of RNA molecules in a cell at a given point of time or the full range of messenger RNA molecules expressed by an organism. The term includes both amount and identity of specific RNA molecules.

Further, with the term “methylome signatures” or “epigenetic signatures” we mean the pattern with which the methylation is distributed in the genome of a cell or an organism when the epigenetics records are restricted to methylation records, and this may also be called a “DNA methylation profile”. The signatures or profiles may be defined in several distribution dimensions: by organ/tissue, by life phase, by genome segment (chromosome), by gene, or by CpG site or by CpG island. The terms “epigenetic signatures”, “methylome signatures” and “DNA methylation profile” have more or less the same meaning since methylation is a major factor of epigenomes and hence the terms are used interchangeably herein.

By the term “gene expression profiles” we mean a collection of a series of transcripts of specific genes as of identity and amount and by the term “transcriptome profiles” we mean expression patterns of the transcriptome studied at cell or specific gene level.

The term “smolt window” is defined as the critical time period where the captive smolt has to be transferred to sea and the wild smolt has to swim downstream and successfully adapt to saltwater to avoid desmoltification; reversal of the smoltification process. Desmoltification will normally cause massive death if transferred under this status. The smolt window is substantially influenced by environmental factors both in culture and in the wild, such as temperature, photoperiod, salinity.

By “smolt status”, we mean status of maturation, readiness for transfer to sea, if evaluated in the smolt window, and general robustness.

By “sea phase”, we mean the period from transfer to sea to harvest and by “sea phase grow-out performance” we mean fish characteristics such as, but not limited to, survival rate, growth rate, specific health or disease records.

By “post-harvest characteristics”, we mean carcass qualities such as meat texture and colour.

By the term “robust smolt”, we mean smolt which in the sea phase displays relative superior records in terms of growth and survival rate.

By the term fish characteristics, we mean welfare traits like fish robustness, health, growth rate, behaviour, appetite, as well as various product qualities like texture and colour. There is a series of subgroups of these characteristics along with the whole lifespan and value chain which may be grouped into welfare needs and welfare indicators both on group, as well as on individual level and guided by The Food Safety Authority and regulated by the Animal Welfare act (in Norway) and similar authorities and regulations elsewhere.

The present invention provides a solution to the problems related to inferior production regimes of farmed fish and particularly to smolt, and also to the problem of traceability of farmed fish, by the use of epigenetics. The provision and analysis of epigenetic signatures of samples of farmed fish according to the method of the invention, can be used in: the verification of fish robustness, health and resulting product quality; in feedback to the aquaculture production, such as the fish farming production, e.g. for amending or optimization of such; as an authenticator or verifier of origin, e.g. to assist in building and protecting brands, or for detecting origin of cultured fish, such as detecting origin of escapees; in vaccinology or other disease preventive measures or in breeding regimes.

The applicant has found that a method, or a process, comprising the provision of at least one epigenetic signature of fish for one or more genes as further listed herein, which may be accompanied with gene expression profiles, can be used as objective biomarker reflectors of fish characteristics, e.g. to achieve any of the above- mentioned applications.

The disclosed and claimed methods take advantage of knowledge and facts about the vertebrate epigenome. The concept of the invention employs epigenetic profiling to identify fish characteristics, such as to further verify and optimize farmed fish production quality. The disclosed method is useful for farmed fish wherein this is from the group of bony fish (teleost). Particularly, the method may be used for teleost fish species that are farmed commercially, and particularly for salmonid species. The fish is e.g. selected from the group comprising Atlantic salmon and brown trout ( Salmo salar and Salmo trutta respectively), Steelhead/Rainbow trout, Chinook salmon,

Coho salmon and other Pacific salmonid fishes ( Oncorhynchus s.p.p.), and is particularly Atlantic salmon. In one embodiment, the fish is from a non-salmonid fish group, such as tilapia, catfish, sea bass or sea bream.

The salmonids go through certain lifecycles, both in captivity and in the wild: When a fertilized egg is ready to hatch, the embryo or juvenile will break free from the egg's soft shell retaining the yolk as a nutrient-rich sac. At this stage, they are called Alevins, or yolk sac larvae. The stage at which the yolk sac has disappeared, and the juveniles have become capable of feeding themselves they are called fry. From start feeding point in time and during an on growing period, depending particularly on management parameters like temperature, they eventually develop into parr with camouflaging vertical stripes. The transformation from parr to smolt is in developmental terms quite fundamental, like a metamorphosis. It is then adapted to a life of drastic contrasts to the existing through vast changes in morphology, salt tolerance, metabolism, behaviour. During this transformation, parr marks fade, fin margins darken, and the body becomes more streamlined with a bright, silvery appearance. In both captivity and in the wild a major behaviour change is the swim pattern, shifting from moving against the stream to swimming downstream. Genetics, feed availability and environmental factors are influencing both time to sexual maturity as well as ocean growth rate.

In fish farming and production of fish in captivity, there is a variety of production regimes, and particularly smolt production regimes. One aim of the inventors has been to uncover epigenetic informative differences between FT and RAS systems and to reveal epigenetic informative markers of and between critical early life phases with emphasis on methylome analysis. In different smolt production regimes the main distinguishing parameters are: temperature and photoperiod structures, the type of system used such as throughflow systems or recirculation systems (RAS), feeding regimes, salt adaptation etc. These parameters also have great impact on the time from hatching or start feeding to smoltification. Based on revealed epigenetic contrasts, either from one type of production regime to another, or from one life phase to another, knowledge about the correlation between epigenetics and fish performance, robustness or health, can be used to adjust the parameters of a production protocol. From the fact that observed reduced robustness and malformations seem to have become an increasing problem in the grow out phase, there is reason to believe that high throughput regimes running with relative high temperatures, such as e.g. over 8 °C at the fertilized egg phase and yolk sac phase, and over 12 °C at the start of feeding of fry, will bestow increased incidences of backbone deformities. One cannot exclude that nature itself has a better solution, leaving the early fragile life in the river at low winter temperature and also leaving the fry to stay and mature for much longer period (e.g. 2-5 years) in fresh water than the commercial regimes which range from less than 1 year to 2 years.

The commercial sea phase grow-out period for Atlantic salmon is currently at an average of 12-14 months until culling at about 5-6 kg. Due to a series of measures over several decades the farming period has been reduced from about 2 years to the current time span. The major measures for this progress are: selective breeding for growth rate and delayed sexual maturation (the latter which otherwise would interfere with growth), feeding regimes, disease control regimes and a series of management.

It is not straight forward to determine the degree of maturation of fish, not least the complex multifactorial smoltification process in salmonid fish accompanied with the challenge of timing of the transfer from freshwater to sea. Also, as described above, sexual maturation, which is unwanted in cost-efficient farming, is a complex trait subject to high plasticity and interaction of genetic and environmental factors. Currently, there are several indicators that may be used in the assessment of maturation, such as smolt maturation, however due to the complex interplay between these, it is not sufficient to test single parameters, such as e.g. analysing the amount of certain proteins or gene transcripts. An epigenetic approach, however, as provided by the method of the invention, will cover necessary and informative biomarkers to reflect maturity and robustness status of importance for optimizing aquaculture of teleost fish, such as in production of farmed fish. Regardless of the fish species, variation in life-cycle, biorhythm, maturation, sex development, being natural or artificially imposed or linked with aquaculture management regimes, all developmental phases and effectors implemented will be reflected in the epigenetic signature, and this is used in the method of the invention.

The applicant provides a method wherein epigenetic signatures are obtained and analysed. In one embodiment, the invention provides a method to identify fish characteristics of farmed fish, comprising a step of preparing at least one epigenetic signature from a fish sample material, wherein the epigenetic signatures are based on selected genes.

In one embodiment, the method comprises the following steps: i) Sampling to obtain fish sample material; ii) DNA Sequencing comprising carrying out genome sequencing, e.g. high through put sequencing, of the fish sample material; iii) Analysing the genome data set of step ii) and establishing epigenetic signatures for the samples; and optionally iv) Comparing and correlating the epigenetic signatures obtained with existing epigenetic signatures, or alternatively or additionally correlating any of the prepared epigenetic signatures with performance data.

The current invention is partly based on findings from studies performed by the applicant. One such study, as further detailed in Example 1 , included the collection of samples of smolt from FT and RAS regimes, DNA isolation and sequencing, generation of epigenetic data, and identification of contrasts in epigenetic signatures. In this study, gene specific epigenetic contrasts between samples from RAS and FT, and further between life phases were recovered by calculating the fraction of methylated CpG sites versus total CpG sites 2k upstream of genes and these fractions were compared between RAS and FT and between life phases. The results are displayed in Tables 1 a and 1 b, of Example 1 , listing top 20, top 100 out of applicant’s data base of top 1000 genes, respectively in terms of differentially methylation levels comparing RAS and FT. Furthermore, Tables 2a-d of Example 1 , list top 80 differentially methylated genes between life phases, distributed with top 20 of each development step from egg to smolt.

The selection of differentially methylated genes, from the comparing of samples from FT vs RAS regimes, were based on the following criteria: Difference in methylation levels, amount of data, biological function of gene product, and to which extent it is expected that gene specific methylation information could assist in optimizing smolt production.

The results of the study firstly show that there are contrasts in methylation levels of certain genes between samples from RAS and FT regimes, respectively. Secondly, the results show that there are contrasts in methylation levels of certain genes between samples from different development stages of smolt, i.e. from egg to larva, from larva to juvenile, from juvenile to parr, and from parr to smolt.

In more detail, the method hence comprises the following steps: i) Sampling steps: The sampling steps include the collection of fish sample material, and this may be from any of individuals, organs, tissues or blood from any of the stages of the fish’ life-cycle which may be used for genome sequencing. The steps comprise the collection of fish sample material, comprising any of early phase whole individuals (eggs or larvae), or selected organs, tissues, or blood from later or several phases of the individual fish. The samples may include fertilized eggs, larvae, fries, parr and individuals at smoltification stages (for salmonid fishes), and ultimately also later from the sea grow out phase until harvest or post-harvest, or organs, tissues, or blood from any of these. The samples may further include material for freezing as well as for RNA stabilisation, the former for methylome analysis and the latter for transcriptome analysis. The fish sample material is hence from any of individuals, organs, tissues or blood from any of the stages of a fish’ life-cycle, comprising either of fertilized eggs, larvae, fry, parr and individuals at smoltification stages, from the sea grow out phase until harvest or post-harvest, or organs, tissues, or blood from any of these. ii) DNA Sequencing steps: The steps comprise carrying out genome sequencing steps of the fish sample material, such as a full genome sequencing, to obtain a genome data set. These steps may include sequencing of the sampled individuals and/or relevant organs, tissues, or blood. These method steps further optionally, but preferably, comprise the step of comparing the obtained genome data set of the fish sample material with an existing genome DNA sequence for the fish. In one embodiment, this is a two-step (bioinformatics assisted) process where the first step comprises alignment of generated sequences with existing reference genomes to pull out annotated genes, followed by a second step where information is obtained about how DNA is methylated within, in the vicinity of, or more distant from certain genes. Reference is made to Example 1. In one embodiment, bioinformatic program software packages, e.g. such as Burrows-Weeler (BWA) and/or NANOPOLISH, are used for these steps of comparing the obtained genome data set with an existing genome DNA sequence, including obtaining information about gene boundaries and information about the methylation. This may further include a step of isolating RNA, either from specific target genes or genome wide. iii) Analysing steps: The steps comprise the analysis of the epigenome, i.e. methylome, of the revealed genome data set and establish epigenetic signatures, such as either or both global genome-wide methylation analysis as well as specific epigenetic signatures. The epigenetic or methylome signatures may be defined in the context of, but not limited to either of: organ/tissue, cell types, genome segment (chromosome), gene and gene related structures, genome regulatory elements (e.g. promoter, enhancer), CpG sites, CpG islands. In one embodiment of the method, the epigenetic signature, and particularly the methylation distribution, of CpG islands is analysed. These may be displayed as e.g. scatter plots, employing contemporary sequencing technologies (e.g. Oxford Nanopore Technologies as described in Example 1), and adequate bioinformatic software both for methylation recognition as well as for displaying the methylation patterns. Additionally, one may also establish gene expression profiles or transcriptome profiles to assist in identification of genes involved together with their level of expression. The methylome signatures and added gene expression profiles may be related to the life phase of the individual, to specific organs, tissues or cells, or to specific candidate genes. Gene expression profiles may be carried out by microarray-based analysis or other platforms (qPCR or direct RNA sequencing) and may be displayed as on/off expression or as quantity of expression, e.g. as heat maps). iv) Comparing and correlating steps: In one embodiment, the established epigenetic signature of step iii) for the fish sample of step i) is compared with existing epigenetic signatures, such as with epigenetic signatures of an epigenetic signature data bank, e.g. as generated by the applicant. Further, this step may comprise the step of correlating any of the prepared epigenetic (methylome) signatures, and optionally the gene expression profiles or transcriptome profiles, with performance data. Such performance data may comprise data for either of:

Life phase, fish (e.g. smolt) management regimes and protocols, traits and performance, sea phase grow out performance, or post-harvest characteristics (carcass qualities). The group of data which the epigenetic signatures is correlated with is herein called “Performance data bank”. The steps preferably include the use of statistical methods and analyses. This may be used for e.g. displaying correlation between different epigenetic/methylome signatures, and/or between epigenetic signatures and fish characteristics such as production protocols and sea phase performance data. When comparing epigenetic signatures with performance data, the method preferably includes a step of statistical correction for different genetic background (e.g. brood stocks), or such correlation should be carried out for fish with the same genetic background.

In one embodiment, the method comprises use of the following groups of data: a) The epigenetic signatures of the fish, and preferably combining this with the genome outputs, e.g. transcriptome or expression profiles; b) Performance data for the fish; e.g. health/welfare and qualities, including e.g. growth rate, survival rate, health records or carcass qualities; c) Production regime data, e.g. from protocols or manuals from the fish producers; d) Observation data from production, i.e. true data from production, e.g. temperature, O2, CO2, salinity, osmolality, or turbidity etc.

The method hence provides a possible merger and comparison of major and critical data relevant for the farmed fish, e.g. combination of any of: Epigenetic signatures and expression profiles; Production regimes and accompanying protocols; and records of various parameters under monitoring.

For the method of the invention the epigenome of the fish is obtained, and this is combined with the genome outputs, e.g. transcriptome and expression profiles, to complete the biomarker structure. Preferably all putative and informative biomarkers of the fish are given room for exposure. The fish biology is hence extensively exposed about its status of welfare opposed to trying to describe it through restricted methods and tests. The method hence includes a sequencing-based whole genome approach.

The method steps may be carried out by employing adequate bioinformatics tools. The obtained epigenetic signatures, as well as results from the expression analyses, may form part of an “epigenetic signature data bank”, as generated by use of the method of the invention. Examples of the sampling and sequencing steps (step i and ii), and analysis step (step iii) together with bioinformatic operations using an adequate set of bioinformatic software, are described in Example 1.

For the sampling steps (i), material is sampled with a minimum sample size (i.e. minimum number of individuals) to ensure fish group representativeness which again is achieved by combining adequate statistics with information on degree of homogeneity of the group subjected to sampling. Preferably stratified sampling is performed, i.e. sampling from a population or a group. In the sampling step of the method, preferably at least one sample material, such as 1-100 fish sample material, such as 1-50 fish sample material, such 2-20 fish sample material, such as 2-10 fish sample material, such as 3-6 fish sample material, such as 4-5 fish sample material, are collected for each data collection. The sampling may further comprise a merger of samples from different individual fish. The sampling strategy should also be taking into account that fish may have different genetic background, e.g. are originating from different breeding regimes, and hence samples should be tagged for such background so as to account for this under step iv).

Further, for the sampling step, and e.g. for the accumulation of data for the epigenetic signature data bank, fish sample materials comprise material from either of:

Different life-phases of the fish’ life-cycle, i.e. preferably from the fertilized eggs, larvae, fish fries, parr or individuals at the smoltification stages, sea grow-out phase until harvest or post-harvest;

Different organs, tissues or cells; e.g. sample material from liver, kidney, brain, guts or gills.

Bioinformatic mining (analysis) of the methylome data may generate the profile or the signatures by genome segments, e.g. chromosomal distribution patterns, by gene in terms of frequency and amount of methylation, and the same applies for selected candidate genes. The genome/gene database of the fish, such as of salmonid species, together with bioinformatic analysis, makes this possible, and the same goes for gene-specific signatures. Flence, one does not have to sample neither chromosomes nor genes but can distribute the methyl signatures on chromosomes and genes using bioinformatics, organ-based methyl data and the salmon genome bank. From this, in one embodiment, for the step of collecting fish sample material, its output signature reflects, either of, but not limited to; different life-phases of the fish’ life-cycle; different organs, tissues or cells; different genome segments, i.e. chromosomes; or selected genes.

Hence, in one embodiment, the method comprises a step of collecting fish sample material, wherein such samples are taken from, or its output signature reflects either of, but not limited to;

Genome: Chromosomal distribution, CpG islands, CpG sites, Gene Associated methylation, or Non-coding methylated areas;

Gene: Gene body, Gene promoter, Gene enhancer, Degree of methylation, or Frequency of methylated genes;

Organ, tissue or cell: Organ/tissue profile (individual organ/tissue, or several organs/tissues collected), Stem cell, Germ line cell, or Cells under differentiation (different phases, e.g. Biological clock at defined cell or defined tissue level), Differentiated cell (e.g. mature B-cells, T-cells);

Life phase (of the organism): Specific phase in life span profile, or the whole life-line profile (e.g. biological clock of the individual).

Referring to Example 1 , and the study of the applicant identifying contrasting epigenetic signature between samples taken from RAS and FT regimes, or from different development stages of smolt, the invention provides a method wherein the epigenetic signatures are prepared from either of the genes selected from the group of those listed in either of Tables 1a, 1b, 2a, 2b, 2c or 2d, provided in Example 1. In one embodiment, the genes are selected from either of those listed in Tables 1a or 1b, and are preferably selected from those listed in Table 1a. In another embodiment, the genes are selected from the group of those listed in either of Tables 2a-d.

Any substantial expansion in smolt production must come from RAS and there is no way RAS protocols can be made equal to FT protocols since the two regimes represent total different approaches. The applicant finds that the only way to optimize RAS is to adjust RAS specific protocols. Currently there is no biological objective criteria to do this based on production information from the more “back to nature like” FT regime. Hence, the solution is to use the “epigenetic language” to learn from FT. Accordingly, identified contrasts in epigenetic signatures, from certain genes, between RAS and FT regimes, may be used to amend RAS protocols. In this way the two regimes also will mutually strengthen each other since also FT regimes can be optimized.

In one embodiment, the invention provides a method employing genes, and preparing epigenetic signatures from these, wherein the genes have a relevant biological function, and wherein there is a difference in methylation levels between samples of FT and RAS. The applicant is in the process of identifying further such relevant candidate genes, wherein there are gene specific epigenetic contrasts between FT and RAS, also assessing to which extent it is expected that gene specific methylation information could assist in optimizing smolt production.

Particularly, the inventors have identified and selected the following differentially methylated genes when comparing FT versus RAS, based on the criteria provided above:

LOC106571646, LOC106601362, LOC106589905, LOC106564914,

LOC106565121, LOC106602814.

Hence, in on embodiment, the invention provides a method wherein the epigenetic signatures are prepared from at least one gene selected from the group of LOC106571646, LOC106601362, LOC106589905, LOC106564914,

LOC106565121 , LOC106602814. As can be observed from Table 1 a of Example 1 , for these genes the “RAS ratio”, i.e. the number of methylated CpG sites divided by the total number of CpG sites per feature for this regime, is much higher than the “FT-ratio” for the same gene.

The selected genes have the following function and industrial and fish welfare relevance:

LOC106571646 actin, alpha cardiac muscle 1-like protein. Highly conserved and involved in cardiac muscle tissue morphogenesis and motility. The sector is observing that FT raised smolt is superior to RAS raised in terms of heart health and the gene appeared a lot less methylated (2 k upstream) in FT than in RAS, suggesting a significantly more activated gene.

LOC106601362 Erythropoietin receptor (EPOR)

The most well-established function of EpoR is to promote proliferation and rescue of erythroid cells (red blood cells) and their progenitors from apoptosis. Hence, an important gene to ensure red blood cell homeostasis and integrity and the gene appeared more active in FT than in RAS.

LOC106589905, BK channel: Potassium Ion channel, Calcium-activated potassium channel, subfamily M subunit alpha-1. BK channels (big potassium), are large conductance calcium-activated potassium channels. They are voltage-gated channels that conduct large amounts of potassium ions (K+) across the cell membrane, hence their name, big potassium. Their function is to repolarize the membrane potential by allowing for potassium to flow outward, in response to a depolarization or increase in calcium levels. They help regulate vital physiological processes like sleep-wake cycle and neuronal excitability. The gene appeared significantly less active in RAS than in FT.

LOC106564914, Cytoglobin-2-like (Cygb2): The gene product is a globin molecule ubiquitously expressed in all tissues. It bestows heme binding, iron binding, oxygen binding and carrier and hence facilitates oxygen diffusion through tissues. It also helps resist hypoxia conditions and serves a protective function during oxidative stress. The gene appears more active in FT than RAS raised smolt.

LOC106565121 , laminin subunit beta-3-like, transcript variant X1. Laminins are a major component of the basal lamina (one of the layers of the basement membranes) a protein network foundation for most cells and organs. The laminins are an important and biologically active part of the basal lamina, influencing cell differentiation, migration, and adhesion. An integral part of the structural scaffolding in almost every tissue of an organism. Laminin is vital for the maintenance and survival of tissues.

LOC106602814, Pro-neuregulin-2, membrane-bound isoform-like.

In one embodiment, the invention provides a method employing either of the genes of Table 1a or 1b, and particularly of Table 1a, and more preferably either of the following genes LOC106571646 actin, alpha cardiac muscle 1-like protein;

LOC106601362 Erythropoietin receptor (EPOR); LOC106589905, BK channel;

LOC106564914, Cytoglobin-2-like (Cygb2); LOC106565121 , laminin subunit beta-3- like, transcript variant X1 ; LOC106602814, Pro-neuregulin-2, membrane-bound isoform-like; with accompanying identified methylation levels, to optimize smolt production regimes. Hence, identified epigenetic information (signatures) from either of these genes, from samples from one type of regime may be used to optimize another regime. More specifically, epigenetic signatures from either of these genes from samples from a FT regime can be used to optimize RAS regimes, such as a guidance to amend RAS protocols. Further, such generated information about contrasts in epigenetic signatures may be used to strengthen both regimes. Based on revealed epigenetic contrasts from one type of production regime to another combined with knowledge about the correlation between epigenetics and fish performance, robustness or health, can be used to adjust the parameters of a production protocol.

Likewise, the inventors have identified and selected differentially methylated genes across life phases (SalmoClock). Out of top 20 differentially methylated genes recovered between each of the four phases (fertilized egg vs yolk sack larvae, yolk sack larvae vs fry, fry vs parr, parr vs smolt), certain genes have been selected based on the following criteria: Difference in methylation levels, genes with rising methylation levels from one phase to next, genes with declining methylation levels, biological function, data volume.

In one embodiment, the invention provides a method employing genes, and preparing epigenetic signatures from these, wherein the genes have a relevant biological function, and wherein there is either a difference in methylation levels between life phases, a rising methylation level from one phase to next, or a declining methylation level from one phase to the next. The applicant is in the process of identifying further such relevant candidate genes, wherein there are gene specific epigenetic contrasts between early development phases of salmon.

Hence, in on embodiment, the invention provides a method wherein the epigenetic signatures are provided from at least one gene selected from either of those disclosed in either of Tables 2a-d, and more preferably from the groups of LOC106613732, LOC106574000, LOC106570740, LOC106610112 (differently methylated from egg to larva);

LOC106609100, LOC106583289, LOC106586831 , LOC106574163 (differently methylated from lava to juvenile);

LOC106565671, LOC106568484, LOC106602974, LOC106560344 (differently methylated from juvenile to parr);

LOC106609239, LOC106609432, LOC106562856, LOC106589831 (differently methylated from (parr to smolt).

The selected genes have the following function and industrial and fish welfare relevance:

For the phases Egg vs Larva:

LOC106613732, Cohesin loading complex subunit SCC4 homolog.

SCC4 is a small (624 amino acids in budding yeast; Saccharomyces cerevisiae) protein containing a multiple-tetratricopeptide-repeats (TPRs) superhelix, whereas SCC2 is a large (1493 amino acid in budding yeast) protein with multiple Huntingtin- elongation factor 3-protein phosphatase 2A-TOR1 repeats. The SCC2/SCC4 complex is specifically required to promote cohesin linkage to chromatin in an ATP- dependent manner at G1/S phase

LOC106574000, transmembrane protein 47-like.

Regulates cell junction organization in epithelial cells. May play a role in the transition from adherent junction to tight junction assembly. May regulate F-actin polymerization required for tight junctional localization dynamics and affect the junctional localization of PARD6B. During podocyte differentiation may negatively regulate activity of FYN and subsequently the abundance of nephrin.

LOC1 06570740, CUB and sushi domain-containing protein 1-like.

Based on analogy to other proteins that contain Sushi domains, it is believed that the gene product of CSMD1 functions as a Complement control protein. Potential suppressor of squamous cell carcinomas.

LOC1 06610112, zinc finger and BTB domain-containing protein 20-like.

May be a transcription factor that may be involved in hematopoiesis, oncogenesis, and immune responses. Plays a role in postnatal myogenesis, may be involved in the regulation of satellite cells self-renewal.

For the phases Larva vs Juvenile:

LOC1 06609100, Cyclin-dependent kinase inhibitor 1B: cyclin-dependent protein serine/threonine kinase inhibitor activity. Cyclins and cyclin-dependent protein kinases (CDKs) are important proteins that are required for the regulation and expression of the large number of components necessary for the passage through the cell cycle. Binds to and stops, prevents or reduces the activity of a cyclin- dependent protein serine/threonine kinase.

LOC1 06583289, insulin-like growth factor binding protein 5 paralog B1.

The encoded protein, mainly expressed in the liver, circulates in the plasma and binds both insulin-like growth factors (IGFs) I and II, prolonging their half-lives and altering their interaction with cell surface receptors. This protein is important in cell migration and metabolism. Low levels of this protein may be associated with impaired glucose tolerance, vascular disease and hypertension in human patients.

LOC1 06586831 , diacylglycerol kinase eta-like, transcript variant X4.

Diacylglycerol kinase that converts diacylglycerol/DAG into phosphatidic acid/phosphatidate/PA and regulates the respective levels of these two bioactive lipids. Thereby, acts as a central switch between the signalling pathways activated by these second messengers with different cellular targets and opposite effects in numerous biological processes. Plays a key role in promoting cell growth. Activates the Ras/B-Raf/C-Raf/MEK/ERK signalling pathway induced by EGF. Regulates the recruitment of RAF1 and BRAF from cytoplasm to membranes and their heterodimerization.

LOC106574163, nuclear receptor subfamily 5 group A member 2-like.

Nuclear receptor that acts as a key metabolic sensor by regulating the expression of genes involved in bile acid synthesis, cholesterol homeostasis and triglyceride synthesis. Together with the oxysterol receptors NR1 FI3/LXR-alpha and NR1 FI2/LXR- beta, acts as an essential transcriptional regulator of lipid metabolism.

For the phases Juvenile vs Parr:

LOC106565671 , AMP deaminase 2-like.

AMP deaminase plays a critical role in energy metabolism. Catalyses the deamination of AMP to IMP and plays an important role in the purine nucleotide cycle.

LOC106568484, inactive carboxypeptidase-like protein X2, transcript variant X2.

May be involved in cell-cell interactions.

LOC106602974, ras-related protein Rab-1 B-like.

The small GTPases Rab are key regulators of intracellular membrane trafficking, from the formation of transport vesicles to their fusion with membranes. Rabs cycle between an inactive GDP-bound form and an active GTP-bound form that is able to recruit to membranes different set of downstream effectors directly responsible for vesicle formation, movement, tethering and fusion. Plays a role in the initial events of the autophagic vacuole development which take place at specialized regions of the endoplasmic reticulum. Regulates vesicular transport between the endoplasmic reticulum and successive Golgi compartments. Promotes the recruitment of lipid phosphatase MTMR6 to the endoplasmic reticulum-Golgi intermediate compartment (By similarity).

LOC106560344, transcription initiation factor TFIID subunit 4-like. Part of the TFIID complex, a multimeric protein complex that plays a central role in mediating promoter responses to various activators and repressors. Potentiates transcriptional activation by the AF-2S of the retinoic acid, vitamin D3 and thyroid hormone.

For the phases Parr vs Smolt:

LOC106609239, aldose reductase-like.

The aldose reductase reaction, in particular the sorbitol produced, is important for the function of various organs in the body. For example, it is generally used as the first step in a synthesis of fructose from glucose; the second step is the oxidation of sorbitol to fructose catalyzed by sorbitol dehydrogenase. The main pathway from glucose to fructose (glycolysis) involves phosphorylation of glucose by hexokinase to form glucose 6-phosphate, followed by isomerization to fructose 6-phosphate and hydrolysis of the phosphate, but the sorbitol pathway is useful because it does not require the input of energy in the form of ATP.

LOC106609432, N-acetylglucosamine-1 -phosphotransferase subunits alpha/beta like.

Catalyses the formation of mannose 6-phosphate (M6P) markers on high mannose type oligosaccharides in the Golgi apparatus. M6P residues are required to bind to the M6P receptors (MPR), which mediate the vesicular transport of lysosomal enzymes to the endosomal/prelysosomal compartment.

LOC106562856, sodium/hydrogen exchanger 5-like.

Involved in pH regulation to eliminate acids generated by active metabolism or to counter adverse environmental conditions. Major proton extruding system driven by the inward sodium ion chemical gradient. Plays an important role in signal transduction

LOC106589831 , serine protease FITRA2, mitochondrial-like Serine protease that shows proteolytic activity against a non-specific substrate beta- casein. Promotes or induces cell death either by direct binding to and inhibition of BIRC proteins (also called inhibitor of apoptosis proteins, lAPs), leading to an increase in caspase activity, or by a BIRC inhibition-independent, caspase- independent and serine protease activity-dependent mechanism. Cleaves THAP5 and promotes its degradation during apoptosis. Isoform 2 seems to be proteolytically inactive.

In one embodiment, the invention provides a method employing genes as disclosed, such as either of the genes of Table 2a-d, and more preferably either of the following genes, found to be differentially methylated across life phases,

LOC1 06613732, LOC106574000, LOC106570740, LOC106610112;

LOC1 06609100, LOC106583289, LOC106586831 , LOC106574163;

LOC1 06565671, LOC106568484, LOC106602974, LOC106560344);

LOC1 06609239, LOC106609432, LOC106562856, LOC106589831 , and preferably adequate statistics, algorithms and machine learning in optimizing the critical early phases of smolt cultivation, such as to construct a dynamic and continuously improvable early phase biological clock (methyl clock) to guide smolt production in harmony with sound fish development and welfare.

Further, the invention provides a method based on any of the forgoing information wherein the prepared epigenetic signatures form part of an epigenetic signature- based test and enhancement system for one or more of the following, but not restricted to, fish welfare or qualities; robustness, maturation, biological age, authentication, in vaccinology and other disease preventive measures or in breeding regimes, of bony fish.

From the method of the invention, providing epigenetic signatures, and optionally gene expression profiles, these may be used for one or more of the following: a) as authenticators for the fish; e.g. for use in traceability of fish. This may include the use as a verification of a given fish production protocol/regime, e.g. for a specific hatchery or smolt or fish farming production regime, such as for use in determination of the origin of escaped farmed fish. b) to distinguish between different production regimes, i.e. different production regimes with accompanying protocols, and further distinguishing between sea phase performance and resulting product qualities from the different production regimes. c) to provide feedback, e.g. to the hatchery and fish farming operators, to assist in optimizing the fish production protocols and regimes, such as the smolt production regimes, and/or the sea phase production regimes. d) to predict sea phase grow-out performance based on smolt epigenetic signatures. E.g. making it possible to distinguish between smolt with different potentials for sea phase grow-out performance, or for predicting the resulting sea phase performance. e) to verify either of quality or origin of the fish, such as to assist in brand building of the fish or smolt. f) to determine or verify the degree of smolt maturation, such as to estimate the timepoint for end of the smolt window, i.e. the timepoint for transfer of the fish to sea water.

Point c) above is a preferred embodiment.

Some, but not restricted to, industrial implications and spinoffs of the invention are:

• Tracking or employed in authentication of fish (since production regimes have unique epigenetic features and hence can be traced)

• Optimizing smolt/salmon production based on the revealed epigenetic contrasts between RAS and flowthrough (FT) and on the methylation based biological clock of a FT regime.

• Optimize RAS “inspired by” FT signatures since FT is regarded as more natural than RAS and FT is regarded by the grow out sector as superior to RAS

• Targeted (gene specific) epigenetic programming experimenting with various production protocols to continuously optimize production using biological and other relevant information from the listed selected genes with differentially methylation status when comparing FT and RAS and when comparing the life phases and when using the methyl-based development clock on the “gold standard” regime FT as guiding tools.

• Employ the above composite epigenetic toolbox as an additional measure to advance genetic gains in breeding and as a systems biology guide at targeting new vaccines or other disease preventive measures.

In one embodiment, the invention provides a method to identify fish characteristics as disclosed, wherein the epigenetic signature is based on either of the genes as identified and selected (any of Tables 1 or 2), and optimizing smolt/salmon production based on at least one revealed epigenetic contrast between either a RAS and a FT system, or between life phases, by adjusting parameters of the production protocol, such as adjusting either of photoperiods, temperatures, feeding regimes and salt adaptation.

Hence, the knowledge obtained from the obtained epigenetic signatures, the “epigenetic signature data bank”, optionally correlated with performance data, the “Performance data bank”, may be used in a verification of the status or quality of a group of fish. Hence, in one embodiment of the method of the invention, a sample is obtained from a fish, at some stage in its life cycle, the epigenetic signature is obtained for this, and this epigenetic signature is compared with existing epigenetic signatures, such as of the epigenetic signature data bank, which for performance data exist, to link this e.g. to environmental conditions, such as to a given regime. Preferably, one should statistically correct for the effect of different genetic background of the fish or carry out comparison of signatures, and signatures and performance within fish of same genetic background.

The applicant has compared the methylome signatures of smolt from different smolt production regimes. When comparing methylome signatures at the smolt window phase as well as gene specific methylation levels between different smolt production regimes (i.e. RAS and Flow through regime, respectively), and corrected for smolt size as well as genetic origin, unique patterns are revealed. In addition, strong contrasts between methylation levels for a series of genes have been found. This implies that the methylome contrasts between the regimes are induced by the measures linked with the regimes and not by genetic origin.

These findings again imply that regime and environment induced epigenetic variation is a novel tool that may be employed, potentially in addition to DNA fingerprint-based traceability, and should not be confused with the latter. Hence, in one embodiment the method comprises testing the methylation status of the genes pointed to herein, and e.g. use this as a quality predictor. The findings from the analysis reported shows that epigenetic signatures can be used as authenticators, e.g. for traceability of fish.

Hence, in one embodiment the method comprises the steps of comparing epigenetic signatures, and preferably additionally also gene expression profiles, from different fish production regimes (seed or smolt or grow out phase) and accompanying environments, and/or comparing such signatures and profiles with performance data, and/or comparing such signatures and expression profiles with production protocols, and/or comparing such signatures and profiles with accumulated databanks of signatures, protocols and performance data.

As a result of such method steps, one would be able e.g. to one or more of:

- Distinguish between different production regimes with accompanying protocols and records and corresponding sea phase performance and product qualities;

- Provide feedback to producers for optimization of regimes and protocols;

- Verify quality and origin of seeds/smolt and farmed fish and resulting products, and hence e.g. assist in building and protecting brands, or to determine origin of escapees.

In one embodiment, the use of such method makes it possible to determine the quality of the fish, such as to distinguish between different fish quality, such as for smolt, farmed fish, such as in predicting the resulting sea phase performance, without assessing single quality parameters, or fish characteristics. In one preferred embodiment the method is for use to verify smolt status, and/or to optimize smolt production quality.

In the method of the invention, the epigenetic signatures can hence be used as a management tool and/or as means for objective biological documentation e.g. for quality. The method may hence form part of a bioproduction. Further, the invention provides a concept using the methylome to provide procedures for both verifying and optimizing the smolt and other seed production and fish farming and the quality of such. The quality obtained is at least in accordance with the requirements of the Norwegian regulations in “The animal welfare act”, “The Food act” and “The aqua culture act”.

The biological/genetics/epigenetics basis of how to achieve these results is based on the knowledge that the epigenome, transcriptome and associated processes provide information about what is going on in the individual at various developmental phases and when exposed to various regimes. This knowledge is accompanied with the employment of the best contemporary technologies available to reveal the natures secrets. Advanced sequencing, translation/bioinformatics and multivariate statistical methods are employed to develop and present relevant profiles and combine and compare them with regime protocols and performance.

For the analysis and comparison/correlation steps, steps iii) and iv) the obtained epigenetic signatures of the sample are analysed and compared with existing data. This may include correlation analysis between either of several epigenetic signatures; the epigenetic signatures and fish performance, robustness or health; or of the epigenetic signatures and production protocols. For the analysis of the epigenetic signatures/epigenome this may include identification of methylation variations and differentially methylated regions, such as of hypomethylation or hypermethylation.

The method hence comprises steps to reveal and identify DNA methylation patterns. The epigenetic signatures obtained, may hence give information of the methylation pattern of regulatory parts of the genome as well as of coding gene regions. The steps preferably comprise the comparing of methylation profiles and the identification of methylation variations. Scatter plots may be generated, which together with adequate bioinformatics and statistics tools will suggest or identify correlations and relationships between the variables, e.g. between different epigenetic signatures, or between a given epigenetic signature and the performance data of the bank from earlier collected samples.

Although there is restricted knowledge about the dynamics of the epigenetic genomic anatomy of fishes, a main part of which is the methylation pattern changes along with development and bio rhythms, a general feature of vertebrates may apply also to most fishes: parental methylation are heavily stripped off during gametogenesis and early embryonic development but also at different modus: first in the male pronucleus as an active demethylation process and later in both parental chromosomes as a passive process during replication and cell divisions. Those parental methylations that overcome the mentioned gametogenic and embryonic reprogramming will represent transgenerational epigenetic inheritance. In addition, a considerable number of genes of vertebrates are differentially expressed in the offspring related to the parent of origin: a copy (allele) of a specific gene inherited from one parent may be expressed whereas the other allele of the same gene inherited from the other parent may be non-expressed. This parent of origin specific expression is called genomic imprinting and is bestowed by parent-of origin differential methylation.

During embryonic phase and later on in development, into adult life and aging, there is an initial re-methylation of CpG sites and mostly none of the CpG islands, followed by both global and organ and gene specific methylation and demethylation for the purpose to take care of normal differentiation and development. The diverse tool package available for methylation and demethylation mechanisms (CpG sites, CpG islands, gene bodies, gene promoters, enhancers etc.) makes the methylome a major instrument in maturation, biorhythms, handling disease and recovery and aging, and in epigenetic responses to environmental effectors and managemental regimes. The rationale behind this is that a series of genes have to come into play and interplay at different phases and influences (either on/off or quantity) whereas most housekeeping genes are on duty on a continuous basis. In parallel with the reprogramming, there is methylome memory established which carry on the whole lifespan of an individual, as explained below as well as transgenerational inheritance as explained above.

There are differential gene expressions depending on the degree of maturation, age and environment. Genes are differentially expressed during an individual’s maturation and aging and this again is governed by endogenous clocks (development, differentiation, biorhythms) and exogenous influences, all of which are released through the main regulator, which is the epigenome, with the methylome as a major contributor. Hence, there is a «biological clock» and a status of maturation reflected through the degree of both global as well as tissue/organ and gene specific methylation, demethylation and expression profiles. In one embodiment of the invention, the degree of gene methylation is used in determining the smolt status, the degree of maturation or smolt quality. Further, in one embodiment, the method is for use in determination of the biological age of farmed fish, or further to determine correlations between biological age and chronological age. Hence, in addition to physiological and behavioural characteristics, the identified fish characteristics may comprise determination of age.

Accordingly, candidate genes, pointed to herein, identified and selected at least partly due to being differentially methylated across life phases (SalmoClock), can be identified playing a putative crucial role at certain life developmental phases or at certain tissue/organ differentiation/specialization phases in addition to the housekeeping genes running on a more continuous basis. Also, these phases and rhythms can be synchronized and accelerated with environmental manipulation like with photo and temperature programs, respectively, e.g. in smolt production. Hence, as part of the sequencing and analysing steps of the method, candidate genes are selected and the epigenetic signatures of these are obtained. These may be compared to methylation information of the databank for the same genes, i.e. the analyses comprise comparing the obtained methylation information of candidate genes with the respective information of the databank for the same genes, e.g. comparing the methylation profiles and identify methylation variations. The methylome or the methylome signature can be studied both at specific points in time (real time) or as a memory signature identifying an experienced regime or environmental impact, which could be both good or inferior.

The methylome has a memory although there is also a continuous reprogramming going on along with the development. This methylome memory pattern or signature can consequently be employed as a reflector of the environmental influences the individual has experienced during its development. Hence, the methylation status of candidate genes will add to both verifying quality as well as to provide more precise feedback to production regimes. Transcripts of such genes or a global transcriptome will add information to the methylome profiles, the former being restricted to time window expression profiles and quantitation without any memory, whereas the latter reflects the regulator landscape and can be memorized, programmed, reprogrammed and inherited.

An individual has its own endogenous biological clock, e.g. for development and aging, and rhythm, e.g. for chronobiology; year, season, lunar, day, night. This is for the major part driven by the methylome, as the methyl groups act as brakes and accelerators and corresponding regulators like hormones. This again can be triggered, accelerated or synchronized, by environment and production regimes, such as e.g. temperature, photoperiod regimes (day and night length etc.). This implies that epigenetic programming can be achieved and managed through environmental stimuli and factors, as an epigenetic exogenous programming. This again means that the method steps of the invention comprising the provision of epigenetic profiling analysis (steps i-iii) along with the correlation steps (iv) wherein the obtained data is e.g. compared with performance data and production regimes, can further potentially include corresponding protocol alterations. This can be employed as a husbandry tool to produce the best possible fish for its purpose. In a preferred embodiment, the method is used to provide the most robust smolt, in optimizing the smolt production, in the preparation of quality smolt, such that in producing high yields and healthy smolt.

Hence, the epigenetic signatures obtained may be used as determinator of fish or smolt maturation, as an “epigenetic clock”. Further, the epigenetic signatures and method of the invention may provide a development status at a certain life phase (Development index). For instance, when comparing groups of smolt from the same operator as well as smolts between operators with different production regimes, the applicant has found that one is able to reveal contrasting difference in methylation frequency, i.e. number of CpG sites methylated compared to CpG sites present in the genome, please see applicant’s PCT/N02021/050030. This methylation frequency parameter may be a useful reflector of maturation since it reflects that different number of genes are in action in the two systems. Hence, given that there is a correlation between number of activated genes and methylation frequency, and this again is correlated to level of maturity and development, the one regime with the lowest frequency of methylation is the one with the most mature fish. Strong evidence for the above being the case came out when blindly analysing 5 groups of smolt within one operator, reference is made to applicant’s PCT/N02021/050030. One group came out with particularly high development index (Group 2), i.e. the inverse ratio of methylation frequency, compared to the others. After testing, the operator informed that the high-level group (Group 2) was given an extra photoperiod treatment during smoltification. Hence, this general development index can be regarded as a particular photoperiod sensitive index and applied as a tool to monitor photoperiod regimes in terms of various structures (protocols) and the effect and outcome of such. Again, this can be used as a tool to optimize photoperiod treatments. Moreover, the development index of groups of smolt from two different operators were analysed. When comparing the two different operators, all groups of the one operator (Operator 1 ) came out significantly higher than the other one in terms of development index. Hence, the provision of epigenetic signatures, and particularly the methylation frequency, may be used in the provision of a development status at a certain life phase. A potential test for methylation level of gene candidates displaying high methylation contrasts between compared regimes is hence useful.

Further, in one embodiment of the invention, a timeline biological clock based on epigenetic (i.e. methyl) signatures is provided for the following salmon development steps: fertilized egg, yolk sack larvae, fry, parr and smolt. This may be used as a Salmon methyl clock, a “SalmoClock”.

The signatures are displayed in several dimensions or by the following features and distribution to allow for maximum informativeness in revealing uniqueness and contrasts between the listed steps of the life phases: chromosome distribution, methyl islands, transcription sites, gene body and 2k upstream gene. Strong contrasts and unique features were revealed for each step both in terms of patterns but also in terms of phase specific genes. The genes were selected based on either of difference in methylation levels, rising methylation levels from one phase to next, declining methylation levels, biological function, data volume. The “SalmoClock”, and the method wherein contrasts in epigenetic signatures between life phases are identified, will provide a robust guidance for safe development of robust smolt, the latter being one of the most critical challenges in current salmon farming.

The epigenome, in contradiction to the transcriptome which has volatile molecules not leaving any trace for memory, has a memory, as described above, and moreover: it can also be inherited through methylome signatures passed to next generation via germ cells. This implies that epigenetics can be deployed to optimize next generation performance through inherited regulatory signals. Also, it implies that epigenetics may be combined with breeding to potentially enjoy a new and untapped synergy. In one embodiment, the method of the invention is combined with breeding, e.g. in epigenetics guided fish rearing. For instance, the fish individuals selected for breeding are picked based on their epigenome, not just their genetic value. In one embodiment of the invention, the analysed epigenetic signature of a fish sample is assessed for its relevance for breeding. Also, fish selected for specific traits and markets could be further finetuned to optimize their performance if epigenetics guided rearing of the smolt and food fish could take place on the top of the current breeding. In one embodiment, the prepared epigenetic signatures form part of an epigenetic signature-based test and enhancement system for breeding regimes.

In the bioproduction context, and in a preferred embodiment of the invention, the methylome signatures, transcriptome or expression profiles (of step iii) could thus be exploited as any one or more of the following:

1) A global dynamic methylation pattern (methylation and demethylation) graph (curve) as a function of development (maturation, differentiation and aging) and thus reflecting an individual’s relative maturation stage or biological age is established.

I.e.: providing the correlation (step iv) between methylation dynamics and maturation/differentiation and aging.

2) A tissue or organ specific dynamic methylation graph reflecting differentiation or maturation is established, to reflect maturation stage and age.

3) As for 1 ) and 2) differentially methylation and/or methylation reprogramming is correlated to maturation/differentiation and aging. 4) As for 1) and 2) differential gene expression (transcripts and transcriptomes) is correlated to differentiation and aging.

Further, the method may include steps wherein any of the results from the correlations steps above are used as maturation and biological age verifiers and as feedback to production and protocols. Main protocols to optimize in smolt production are light (photoperiods) and temperature regimes. Inferior maturity in smolt production may therefore mostly be related to the structuring of these parameters. In one embodiment, the obtained epigenetic signatures and/or gene expression profiles can be linked with the sea phase performance, e.g. if this is good or inferior.

Along with the accumulation of data for the epigenetic signature data bank and the performance data bank, the method of the invention and value and usefulness of this, will gain strength. Hence, along with the accumulation of global or organ/tissue/cell or gene based epigenetic signatures and expression profiles linked with life phase, and correlated with production regimes/protocols and grow out sea phase performance data, and statistical association calculations between such, the method, both for verification and feedback, will gain strength. In one embodiment, the method requires big data compilation and eventually also likely machine learning (ML) implementation.

Hence, the method may include the use of tools and methods to handle informatics, bioinformatics, statistics or mathematics, which may comprise any one or more of the following, but not being restricted to:

Image and pattern analysis and recognition (e.g. scatter plots), cluster analysis, various comparison and probability biostatistics within regression analysis (least squares, linear and non-linear), multivariate analysis and data dimensional reduction techniques, fish index calculations based on signatures, computer graphics, big and large scale data, machine learning and artificial intelligence techniques to handle complex and vast data together with adequate algorithm and computer software development and/or customization.

Genome location: The 2k upstream of genes is a key genome location of methyl signatures with powerful universal informative value. Analyses of genome wide methylation levels, restricted to those 2k upstream of gene reading frames, display variability far above all other genome location or features when comparing salmon life phases from fertilized egg to smolt as well as comparing different production regimes at smolt window phase. This finding implies that this 2k upstream “methyl universe” is the most robust source from which to find informative methyl patterns as well as concrete gene specific methylation levels linked to, but not restricted to, origin (authentication and traceability) as well as to maturation (unwanted maturation included like sex maturation in the grow out phase), development status, robustness etc.

In one embodiment, the method comprises that any one or more of the above- mentioned candidate genes or groups of genes are selected and the epigenetic signatures, the methylation status, and optionally also expression profiles of these are obtained. In one embodiment, the method comprises preparing at least one epigenetic signature for one or more of the genes identified herein, such as of anyone of Tables 1a, 1b, 2a-d.

Further, in one embodiment of the invention, the prepared epigenetic signatures may form part of an epigenetic signature-based test and enhancement system for either of fish welfare and qualities, such as of either of robustness, maturation, biological age, authentification, in vaccinology and other disease preventive measures or in breeding regimes.

Based on the findings (observations) from the identified epigenetic signatures and optional gene expression profiles (steps iii), and from the results of the correlations steps (steps iv) of the method, adequate measures, i.e. additional potential steps of the method, that can be taken are indicated below:

Observation: The general (global) demethylation and/or differential methylome and transcriptome is not satisfactory advanced compared to life phase, i.e. is immature. Measure; Implement extended photoperiod or optimized day and night regime and leave more time. Observation: Tissue/organ specific differentiation is not satisfactory developed, including chloride cells.

Measure: Extend or re-structure photoperiod with e.g. strengthen “winter modus” (see paragraph below) and/or extend time for maturation with accompanying lower temperature.

Observation: Gene specific methylome and gene specific expression profile related to smoltification is not in place. I.e. «the freshwater hormone» prolactin should be downregulated, thyroxine, NA-K-ATPase and mineral corticoids, i.e. cortisol («the salt-water hormone») should be upregulated, 02-sensitive haemoglobin variants should be upregulated due to preparing for lower oxygen tension in seawater etc. Measure: Change day/night ratio to initially have shorter days (winter period) before extending day period to synchronize the various preparing processes in the fish. Consider exposing the smolt to more salinity and lower 02-tension for a defined period to stimulate chloride cell development and haemoglobin variant switch.

Observation: Sub-optimal performance in the grow out sea phase and post-harvest qualities.

Measure: Compare through advanced statistics the performance data with smolt production regime/protocols and with smolt methylome signatures and expression profiles. Depending on type of sea phase malperformance (early, mid phase or late death or sickness, cause, carcass qualities) and correlation analysis results with profiles and smolt regimes/protocols: sort out targeted feedback to alter production protocols to the better. For instance, if the grow out records show inferior survival rate due to infections and the epigenetic signatures and expression profiles of the corresponding smolt reflects inferior maturation and differentiation of immune organs and tissues, the feedback to hatchery protocols should be to leave more time for the fry to mature and/or to optimize feed formula. The method comprises the steps of correlation analysis between either of epigenetic signatures; epigenetic signatures and fish performance, robustness or health; or epigenetic signatures and production protocols. The provision of the epigenetic signatures and the correlation analysis may be used in the verification of fish robustness and health and resulting product quality; in feedback to the fish farming production; as an authenticator and verifier of origin e.g. to assist in building and protecting brands; or for detecting origin of cultured fish, such as detecting origin of escapees.

Existing tracking systems based on brood stock and pedigree information have a restriction on assigning to locations and production regimes opposed to epigenetic profiles which are strong reflectors of such.

A series of parameters can be adjusted depending on if the plant is a flow through system or a recirculation aquaculture system (RAS). The major common parameters are:

Photoperiod regime (day/night ratio) and time, temperature and time. In addition to these, both regimes can alter: Feed and feeding regimes, including parameters as water flow, fish density, oxygen and CO2 concentration, salinity exposure, handling regimes (e.g. moving fish to new compartments along with growth).

Hence, in one embodiment of the method, fish farming production regimes and corresponding protocols are adjusted or amended, such as optimized, based on feedback from the prepared epigenetic signatures, optionally linked with the performance data.

Examples of appropriate procedures, tools and instruments to use in the method are provided herein. For the sampling and sequencing steps, analytical methods are to be used, e.g. including DNA isolation from relevant samples, e.g. from smolt organs and tissues. Genomic DNA is isolated from relevant organs, e.g. from Atlantic salmon ( Salmo salar ) using standard, well known protocols. E.g. DNA is isolated using the kit “DNeasy Blood and tissue” kit (Qiagen) following the recommended protocol. Shortly: A suitable amount of biological material from the fish tissue or organ is digested by proteinase K and corresponding buffer. The solution is mixed with relevant kit buffer and ethanol and centrifuged through a spin column where the DNA is bound to a filter with affinity for this. Appropriate kit buffers are used for washing and eluting the DNA from the spin column.

Alternative methods for isolating genomic DNA from the fish samples may be used, e.g. phenoLchloroform may be used to extract proteins and other molecules after the proteinase K treatment. DNA may then be precipitated using ethanol and salt.

The quantity and purity of the DNA may be recorded using e.g. Qubit and Nanodrop instruments, respectively.

DNA sequencing: Genomic DNA is sequenced e.g. using the MinlON instrument from Oxford Nanopore Technologies (ONT) and associated sequencing kit: Ligation Sequencing Kit (SQK-LSK109). The recommended protocol of the producer was followed. Shortly: The genomic DNA is treated with kit ingredients to repair the ends of the molecules as well as generating a 5’ A-overhang. Sequencing adapters, containing the necessary DNA sequence and molecules in order to guide the DNA molecule through the pores in the flow cell membrane of the MinlON instrument, are ligated to the genomic DNA fragments. The DNA molecules are then loaded onto the flow cell and the sequencing process runs until a satisfactory number of DNA sequences are produced. Both R9.5.1 Flow cells as well as Flongle flow cells may be used, but the R9.5.1 MinlON flow cell has the preferred high capacity when whole genome sequencing is performed. The MinlON sequencing instrument from ONT is used. Flowever, other instruments from the same providers (GridlON and PromethlON) operating with the same DNA sequencing technology as MinlON, but with higher throughput capacity, may also be used. The sequencing process is controlled with the accompanied software MINKNOW provided by ONT.

RNA isolation: RNA may be isolated from relevant fish sample materials, such as from smolt organs and tissues.

The epigenetic fingerprint (signatures) obtained, e.g. as described in the Examples, give information of the methylation pattern of regulatory parts of the genome as well of coding gene regions. These methylation patterns affect the expression of genes that have influence on important traits of the fish - both production traits as well as other traits and performance characteristics. This gene expression influence may be in the form of increased or decreased transcription of specific genes. Thus, the amount of mRNA is affected. In some situations, it may be of interest to analyze the relative amount of specific mRNA molecules present in the fish sample material, such as in relevant organs and tissue of fry and smolt as well as adult individuals, and the method may comprise this.

Total RNA is isolated using standard techniques. Samples from relevant organs and tissue from fish is placed on RNA preserving buffer (RNAIater) or immediately frozen in liquid nitrogen. Total RNA and/or mRNA is isolated. Either by spin column methods as Qiagens RNeasy Mini Kit or similar kits from other providers. Also, traditional methods based on Trizol extraction of proteins works perfectly.

RNA sequencing and/or quantification (Nanopore or qPCR, stabilized on RNA Later stabilization solution): Isolated total RNA or mRNA can be sequenced using several sequencing methods. For example, RNA molecules may be sequenced directly using the Oxford Nanopore kit “Direct RNA sequencing kit” and the same instrument as described above, MinlON, for direct DNA sequencing. By sequencing the RNA molecule directly, base modifications in the molecule can be detected. The isolated RNA molecules can also be transcribed to cDNA which subsequently are sequenced using either the Nanopore technology or other available techniques and instruments.

In one embodiment, the method comprises a step of transcriptome analysis based e.g. on data from microarray, Nanopore based sequencing or qPCR: Bioinformatics Statistical analysis displaying correlation between methylome signatures, gene expression profiles, production protocols and sea phase performance data, such as contract biostatistics and programmer expertise.

To summarize, the main concept of the invention comprises the preparing and comparing of epigenetic signatures, as displayed in Example 1 and above, from certain genes. The epigenetic signatures may be compared with fish production regimes (protocols) and preferably with performance data records (growth rate, health and carcass qualities). Performance on a continuous basis may be recorded consisting of, but not limited to, growth rate, survival rate, health records as well as carcass qualities. Samples may also be taken from several relevant organs/tissues of culled fish as basis for generating epigenetic signature of mature fish.

Hence, by employing contemporary biostatistics or informatics tools in data mining, compiling and comparison, this enables the provision of either of the following major solutions:

• Distinguish between different production regimes, i.e. different production regimes with accompanying protocols, and further distinguishing between sea phase performance and resulting product qualities from the different production regimes.

• Provide feedback to optimize production through adequate alteration of protocols guided by epigenetic signatures.

• Provide prediction of performance of fish in the sea phase based on smolt epigenetic signature.

• Establish epigenetic signature-based tracking of fish to assist in: authenticity, brand building and brand protection of seafood and detection of origin of escaped farmed fish. This will be based on the establishment of signature databases of various producers with which signatures of seafood products in the markets or living escapees will be compared.

The following Examples are provided to illustrate the invention in accordance with the principles of the invention but are not to be construed as limiting the invention in any way.

Examples

Example 1. Preparation and analysis of epigenetic signatures from samples from RAS and FT reqimes

Sampling and preservation:

60 organ samples were collected from each RAS and FT regime, i.e. 4 organs (gills, anterior and posterior kidney and liver) from 15 randomly selected smolt, of Atlantic salmon, from each of FT and RAS regimes. The smolt were anesthetized in accordance with animal welfare regulations and preserved on dry ice immediately and kept as such until frozen at minus 80 °C in the lab. Smolt from both RAS and FT represented the same genetic background (same genetic broodstock) to make sure that there would be no genetic bias on the epigenetic results. Also samples from the various life phases from egg to smolt “The SalmoClock” originated all from the same genetic broodstock to ensure not confusing epigenetic life phase contrasts with differences in genetic background. The SalmoClock material was sampled from a FT regime. For the SalmoClock-material the whole organism represented the sample for the life phases egg, yolk sack larvae and fry whereas for parr and smolt the same organs as for the regime experiment represented the samples. Since FT is a more well proven and mature regime than RAS in addition to being closer to the natural environments of salmon, it is expected that a FT based biological early phase methyl clock will be of high value to “train” and optimize the RAS regimes.

Preparation of samples and DNA isolation:

Genomic DNA was isolated from the organs and from the whole individuals of the earliest life phases: fertilized eggs, yolk sack larvae and fry using standard, well known protocols e.g. DNA was isolated using the kit “DNeasy Blood and tissue” kit (Qiagen) following the recommended protocol. Shortly: A suitable amount of biological material from the fish tissue or organ was digested by proteinase K and corresponding buffer. The solution was mixed with relevant kit buffer and ethanol and centrifuged through a spin column where the DNA was bound to a filter with affinity for this. Appropriate kit buffers were used for washing and eluting the DNA from the spin column. In the part of the study comparing RAS and FT regimes, equal amounts of DNA were pooled from each organ such that also an equal amount of DNA was produced from each regime under comparison. For the part of the study comparing epigenetic signatures of different phases, the “SalmoClock experiment”, DNA from 100 whole individuals of fertilized eggs, yolk sack larvae and fry was pooled, representing each phase whereas for the parr and smolt phase the same organs as for the FT/RAS regime experiment were selected and pooled to represent these phases.

DNA sequencing:

Genomic DNA was sequenced e.g. using the MinlON instrument from Oxford Nanopore Technologies (ONT) and associated sequencing kit: Ligation Sequencing Kit (SQK-LSK109). The recommended protocol of the producer was followed.

Shortly: The genomic DNA was treated with kit ingredients to repair the ends of the molecules as well as generating a 5’ A-overhang. Sequencing adapters, containing the necessary DNA sequence and molecules in order to guide the DNA molecule through the pores in the flow cell membrane of the MinlON instrument, were ligated to the genomic DNA fragments. The DNA molecules were then loaded onto the flow cell and the sequencing process run until a satisfactory number of DNA sequences were produced. Both R9.5.1 Flow cells as well as Flongle flow cells may be used, but the R9.5.1 MinlON flow cell has the preferred high capacity when whole genome sequencing is performed. The MinlON sequencing instrument from ONT was used. Flowever, other instruments from the same providers (GridlON and PromethlON) operating with the same DNA sequencing technology as MinlON, but with higher throughput capacity, may also be used. The sequencing process was controlled with the accompanied software MINKNOW provided by ONT. Basecalling was carried out with GUPPY, which generates the fastq sequence files. The average sequence data depth was between 1 ,5x and 2x genomes.

Generation of epigenetic data:

Epigenetic data and contrasts between RAS and FT as well as between early development phases, were generated using adequate bioinformatic tools and algorithms/statistics.

Generation of global, organ and gene specific methylation profiles: The reads were then aligned to the Atlantic salmon (Salmo salar) reference genome using Burrows-Weeler (BWA) software package. Methylated CG sites were recovered using the Oxford Nanopore software package, NANOPOLISH, for signal level analysis to detect methyl modifications, here 5-methylcytocine of CG sites of the sequence data. Instances with a statistical log likelihood > |2.5| were considered a valid instance.

Then, using the script, “calculate frequency”, each CpG instance was cumulatively summed, so that each CpG site had two numbers: number of reads covering the site and number of times that site was called methylated. The average global (whole genome) CpG and methylation instance data depth of regimes and life phases under comparison before narrowing the data to various genome features (see below) was approx. 40 and 30 million, respectively, representing an average global methylation fraction of approx. 70%.

Adequate bioinformatics was used to select and study methylation events on the following genome features: global (10 and 100 k window), gene body, islands and 2k upstream. As displayed in Figure 1 , 2k upstream came out with the greatest variation of methylation across genome features. Consequently, this genome feature was also used to study contrasts between FT and RAS as well as life phase contrasts both at pattern/global and gene specific levels.

Hence, the gene specific epigenetic contrasts between RAS and FT and between life phases were recovered by calculating the fraction of methylated CpG sites vs total CpG sites 2k upstream of genes and these fractions were compared between RAS and FT and between life phases. The results are displayed in Tables 1a and b listing top 20, and top 100, out of our data base of top 1000 genes, respectively in terms of differentially methylation levels comparing RAS and FT.

Below Tables 1a and 1 b hence provide a ranked list of top 20 and top 100, respectively, selected genes using a window of 2000 bp upstream region for each predicted gene. The gene list is ranked by absolute differences (ratio minus ratio) comparing two different production regimes: flow-through and RAS regime. The ratio value is calculated from the number of methylated CpG sites divided by the total number of CpG sites per feature. Both regimes were standardized in terms of genetic background (same broodstock) and both were smolt matured up to the smolt window and comprised smolt of same size. The columns listed from left to right: gene, number of CpG sites of this gene of the RAS regime, number of methylated CpG sites of this gene at the RAS regime, number of CpG sites of this gene of the Flow through (FT) regime, number of methylated CpG sites of this gene of the FT regime, ratio of methylation of this gene of the RAS regime, ratio of methylation of this gene of the FT regime. Furthermore, Tables 2 a-d below, list the top 80 differentially methylated genes between life phases, distributed for the top 20 genes of each development step from egg to smolt. Flence, this is a ranked list of top 20 selected genes using window of 2000 bp upstream region for each predicted gene. The gene list is ranked by absolute differences (ratio minus ratio) comparing pairwise life phases of salmon. The ratio value is calculated from the number of methylated CpG sites divided by the total number of CpG sites per feature. Columns listed from left to right: gene, number of CpG sites of this gene of a specific life phase, number of methylated CpG sites of this gene of the same life phase, number of CpG sites of this gene of the compared life phase, number of methylated CpG sites of the compared life phase, ratio of methylation of this gene of the first life phase, ratio of methylation of this gene of the compared life phase. Table 2a provides the top 20 differentially methylated genes from egg to larva. Table 2b provides the top 20 differentially methylated genes from larva to juvenile. Table 2c provides the top 20 differentially methylated genes from juvenile to parr. Table 2d provides the top 20 differentially methylated genes from parr to smolt.

Table 1a:

Table 1b:

Table 2a:

Table 2b:

Table 2c: Table 2d: