This guest post is by Ewan Birney on Genomic and phenotypic characterisation of a wild Medaka population: Establishing an isogenic population genetic resource in fish, arXived here.

Our lab is part of a collaboration spanning Japan, two groups in Germany and EMBL-EBI in the UK which put the paper “Genomic and phenotypic characterisation of a wild Medaka population: Establishing an isogenic population genetic resource in fish” on to the arXive pre-print server. Here I’ve taken up a kind invitation to post about this paper.

Before getting into the details of the paper, I should introduce Medaka, scientific name, Oryzias latipes. Medaka is a small fish which lives in both fresh and brackish water across the majority of the Japanese archipelago (the exception is Hokkaido, the large northern island in Japan), the Korean peninsular and the eastern China region. It has been kept as a garden pet (and then in aquaria) in Japan for a long time, with documented colourful strains in Japanese art since 17th Century (this is a woodblock from the celebrated artist, Ando Hiroshige, (安藤 広重) depicting Goldfish and, the much smaller, Medaka fish – the medaka fish are the small horizontal shoals, not the colourful goldfish sadly). It is widespread in the wild, in particular in rice paddies, hence its other common name, “the Japanese rice paddy fish”. After the rediscovery of Mendel’s work at the turn of the twentieth century scientists in Japan started to use the established colourful strains to explore genetics. The most famous paper from this era is the first discovery in any species of crossover of the sex chromosomes, X-Y. Rather brilliantly this is an open access paper from Genetics (Aida T: On the Inheritance of Color in a Fresh-Water Fish, APLOCHEILUS LATIPES Temmick and Schlegel, with Special Reference to Sex-Linked Inheritance. Genetics 1921, 6:554-573. http://europepmc.org/articles/PMC1200522 Note: at the time, the systematics in this area of fish was different, hence the different genus name). Since this early genetics, Medaka has been used for research both inside and outside Japan over the 20th Century, with a well established linkage map, transgenic procedures, genome sequence, and considerable probing of different phenotypes.

My experience in describing Medaka to European and American audiences is to now answer the usual questions about similarity and contrasts with Zebrafish, the most commonly used laboratory fish in the West. The first important thing to realise is that Medaka is on a very different branch of the Teleost (bony fish) lineage from Zebrafish, separated by an estimated 250-300 million years of divergence – so the Medaka fish genome is only marginally closer to the Zebrafish genome than either are to mammals. One should expect quite different biological details in the two systems, and each system to be equally applicable to mammalian systems. Medaka is somewhat smaller than Zebrafish and nearly always will live in the same tank format as Zebrafish and the same water system (many labs co-culture both Medaka and Zebrafish). Generation time is similar (6 weeks ~ 3 months). Zebrafish lay around 1,000 eggs in a single mating, which is a distinct advantage to Medaka’s clutch of around 30 eggs in a single mating, held to the female. However whereas zebrafish mate only once per week requiring a ‘recovery phase’ , medaka mate every day. Thus the difference in fecundity over time is small. Both zebrafish and medaka have transparent chorions (egg shells); also the embryo itself is completely transparent in both species rendering them ideal model systems to study development. Generally all techniques that have been established for one species are also applicable for the other, such as transgenesis by simple injection of suitable DNA vectors or antisense morphlinos. Medaka fish genetics is cleaner, with many inbred laboratory lines maintained by single brother-sister mating, and thus very homozygous throughout. Finally these medaka inbred lines are often made from wild-catch individuals, with an established breeding protocol to achieve homozygosity from wild individuals.

It is this last feature that we would like leverage here. With an inbreeding protocol from the wild, one can set up a near-isogenic wild panel, similar to the panels that have already been developed in Arabidopsis and Drosophila. These panels are proving very informative for quantitative genetics in both of these fields. Once such a panel is established it is a powerful mapping resource and is one of the few ways to study gene-environment effects as one can repeat phenotyping experiments over the same genotypes but in differing laboratory environments (say, high to low calorie diets, or different temperatures, or different small molecules added to the water). Being able to have such a panel with a vertebrate will be very powerful (Medaka fish has all the common cell types and tissues for a vertebrate – brain, heart, liver, muscle, gut, pancreas – both endocrine and exocrine, kidney.) The main purpose of this paper is to find and characterise the source wild population for such a panel.

A good genetic panel needs ideally to be free of population structure. One also needs the right linkage disequilibrium (LD) properties. In previous decades there was a sweet spot for LD in the genome; the longer the LD the cheaper it was to genotype, but shorter LD gave better resolution of where a functional variant is. With the advent of cheap sequencing, this trade off logic has changed to finding a population with short LD to have the best possible mapping resolution. Finally there are also practical aspects for choice of population – one would like to be able to resample from the same region easily (for example, to add to the panel in the future). After looking at a number of sites where we assessed population properties via mitochondrial typing, we choose a site, Kiyosu (https://maps.google.com/maps?q=34.78113,+137.347928333056(Kiyosu%20Sampling%20Site)&iwloc=A&hl=en), close to Nagoya where the NIBB Medaka resource group under Kiyoshi Naruse is housed. From this population we caught a number of individuals and set up 8 breeding pairs. For each of the 8 pairs, we sequenced the two parents and one child (a “trio” in the parlance of genetics). We choose this sequencing structure as it means we can phase the parental genotypes using the child’s genotype, and in effect sample 16 haplotypes from this population.

From this we show we have a good population for an isogenic panel. There is no discernable population structure, both from a distance matrix perspective across all individuals, and the lack of long LD in the population. As expected for a large teleost population, there are a relatively large number of variants, with a segregating SNP every 150bp on average. In this limited sample the LD is, as expected, quite tight, with the correlation between SNPs (expressed as r2) dropping to “baseline” levels between 5 – 10 KB . From even this limited panel we estimate that almost 40% of SNPs would be mappable to a single exon. We expect this will improve in a more complete panel.

To augment this population characterisation, we also performed some high throughput phenotyping, showing that the population has expected phenotypic diversity driven by genetics. To do this we took advantage of a small number of existing inbred southern line strains. As the numbers are low here, we cannot map any phenotypes (this will require the panel), but we can get an estimate of broad sense heritability, i.e., the proportion of variance of a measurement explained by the differences between inbred lines compared to the differences between individuals of the same line. This is the sort of calculation which is easy to do on inbred panels, and harder to deconvolute on family or outbred cases. For 6 out 7 traits we chose to measure we get reasonably high broad sense heritability measures.

We conclude that we have a good source wild population that is likely to lead to a successful near isogenic panel, and have started inbreeding. We are somewhere between the 3rd to the 4th round of brother-sister inbreeding for 200 founder pairs, and all the lines look healthy. Traditionally one considers lines to have inbred after 8 brother-sister matings, so we’re almost half way through. Although in theory the generation time is 3 months, when one does this on 200 lines, the logistics means it is closer to 4 to 5 months. Felix Loosli is overseeing the inbreeding at the KIT (Karlsruhe Institute for Technology), in Germany.

In addition to the main thrust of this paper, we also look at the population genetics of Medaka, as this is a large wild catch with complete genome wide coverage of SNPs. For me the most interesting thing is the relationship with the Northern strain of Medaka. Japan has a large mountain range roughly running down the middle of the main island of Japan (called the “Japanese Alps” in English –日本アルプス Nihon Arupusu in Japanese). The Medaka fish found to the north of this are phenotypically different (more heavily pigmented, and prefer to live in shallower tanks), but do interbreed in the laboratory with Southern strains. An open question is whether there has been any partial interbreeding (called introgression) in the wild. Using sensitive tests for introgression between lineages, first developed to detect the Human/Neanderthal interbreeding, we do not see evidence for wild interbreeding between the Northern and Southern strains. This lends support that these two “strains” are best thought of as separate species, and one might expect there to have been specific selection events for the phenotypic properties in both strains.

We welcome comments on this paper, but also more generally on the use of the future Kiyosu near isogenic wild panel. We will be completely open about data collection and distribution of data for this panel, wherever possible using global archives to minimise the complications in getting access to the data.

If you are a teleost biologist, the fact that one can co-culture Medaka and Zebrafish means that many phenotypic assays set up for Zebrafish are probably quite easily transferable to Medaka; there is an existing (small) set of inbred lines which you could look to develop assays on. If you are a molecular quantitative geneticist, this panel should have better mapping properties than human or mouse, but a full range of vertebrate cell types. I am looking forward to using some of the statistical techniques developed on Arabidopsis and Drosophila, in particular the gene/environment partitioning components, and if you are interested in gene/environment interactions, this panel will have some unique opportunities. Of course, this is not a panacea – it takes investment in husbandry details and logistics to bring in another species into any system, and quantitative genetics is just one way of exploring genetic effects – forward and reverse genetics are very powerful techniques (which, incidentally have both been used in Medaka).

If you are interested do contact Ewan Birney (birney@ebi.ac.uk) or Ian Dunham (dunham@ebi.ac.uk) on aspects of the genomics/variation, and Felix Loosli (felix.loosli@kit.edu) , Jochen Wittbrodt (Jochen.Wittbrodt@cos.uni-Heidelberg.de) or Kiyoshi Naruse (naruse@nibb.ac.jp) on Medaka husbandry and molecular biology.

This guest post is by James Lee on his paper with Carson Chow, “The causal meaning of Fisher’s average effect“, arXived here

Early in graduate school, I took it upon myself to read Reinhard Burger’s excellent treatise The Mathematical Theory of Selection, Recombination, and Mutation. Here I encountered the concepts of “average excess” and “average effect,” which were defined (rather unclearly to the casual reader) by Ronald Fisher in his presentation of the Fundamental Theorem of Natural Selection. Finding some of the distinctions made between these two concepts rather confusing, I directed some questions about them to the Yahoo quantitative genetics group. A respondent told me to consult Falconer (1985), which would “make things as clear as mud.”

My school did not have electronic access to Genetics Research at the time, so I did things the old-fashioned way and got my hands on a bound copy of the journal volume containing Falconer’s article. This masterpiece of exposition impressed me so much that I copied it down by hand; since the paper was at the end of the bound volume, the librarian was not able to scan it for me.

Falconer set out four distinct concepts that at various times have been put forth as definitions of the average excess, average effect, or both:

(A) Divide the population into two groups, one containing all A1A1 homozygotes and half of the heterozygotes, the other containing all A2A2 homozygotes and half of the heterozygotes. Take the difference between the conditional mean phenotypes of these two groups.

(B) Choose gametes bearing A1 and A2 at random. Measure the phenotypes of the mature organisms to which these gametes ultimately give rise. Take the difference between the conditional mean phenotypes of the A1 and A2 gametes.

(C) Regress the phenotype on the count (0, 1, or 2) of an arbitrarily chosen allele (A1 or A2). Take the regression coefficient of gene count.

(D) Take the average change in phenotype resulting from experimentally “zapping” one allele into the other, as if by mutation, in a zygote immediately after fertilization but before the onset of any developmental events.

Implicitly assuming that genotypes and environments are independent, Falconer then showed that all four concepts are equivalent under random mating. Now suppose that mating is not random. Then (A) and (B) are still equal and correspond to what Fisher called the average excess. The numerical value of this quantity is generally not equal to either (C) or (D), and in turn (C) and (D) are generally not equal to each other. Falconer concluded that (C) was what Fisher really meant by the average effect.

This conclusion disturbed me a great deal. As any GWAS researcher knows, the (partial) regression of phenotype on gene count does not necessarily pick out any biologically meaningful quantity if genotypes and environments are dependent (“population stratification”). The fundamental issue here is that (C) is merely a statistical definition, appealing only to passive observations of a static population, whereas (D) is a causal definition turning on the result of a hypothetical experimental intervention. I no longer remember now whether I had read Pearl (2009) by this point, but regardless my Spider Sense was unambiguously telling me that (D) was deeper and more meaningful than (C). Furthermore, if Fisher was not the one who coined the slogan “correlation is not causation,” he was certainly one of its first and most vocal promoters. How could Fisher, who invented randomization in experimental design, have preferred a correlational definition over a causal one when setting forth one of the key concepts in his evolutionary theory? Could it be because of the difficult in translating (D) from words into mathematical symbols without something like Pearl’s do operator, which was not available in Fisher’s time?

This paradox continued to bother me over the next several years. Soon after my daughter was born, I indulged one of those wild impulses that strike the sleepless: I emailed my questions regarding this matter to Anthony W. F. Edwards, the last student of the great Fisher himself. Anthony very generously sent me some of his unpublished work and also his correspondence with Falconer about the very article that had spurred my thoughts. This correspondence spanned a period of more than 20 years, and it provided a very poignant portrait of Douglas Falconer as a scientist (Hill and Mackay, 2004). I did not immediately find the answers to my questions in the materials that Anthony sent to me, but they set me on the path toward finding the answers. These are presented in the paper, which will shortly appear in Genetics Research.

It turns out that Fisher’s average effect must be given a causal interpretation after all. For the detailed story of the reconciliation between (C) and (D), you will have to read the paper, written in collaboration with my supervisor Carson Chow. I am particularly pleased with our proof that the frequency-weighted mean of the (experimental) average effects at any locus is equal to zero. In most texts this relation is extrinsically applied to the multilocus case without any motivation except that it holds automatically for the (regression) average effects in the case of a single locus. The fact that this identity, which otherwise is an arbitrary constraint, can be derived from a definition positing the experimental replacement of a homologous gene is rather striking evidence for the importance of a causal interpretation.

Our investigation unexpectedly turned up many connections to other parts of population genetics. I like to think that in the pages of our paper one can hear many masters of population and quantitative genetics–Hardy, Fisher, Wright, Kimura, Falconer, Price, Ewens, Lessard–engaging in a deep conversation.

There are some issues raised in the paper that I am still contemplating. First, there is a complication when one considers randomly sampling a zygote and experimentally changing its genotype to the one whose value needs to be known; such an experiment inevitably changes the frequencies of the genotypes, and for theoretical reasons any ensuing frequency-dependent changes in the phenotypic means of the genotypes needs to be excluded. I believe that one way to do this properly is by partition of the effects of the experiment according to Wright’s path analysis–which would be rather ironic given the well-known antagonism between Wright and Fisher. Second, in the multilocus case it might be possible to mathematically describe special subsets of possible gene substitutions defining a given average effect that satisfy the property that all changes in Hardy-Weinberg and linkage disequilibria are “small.” We look forward to future work (by ourselves?) on these questions.

Note: The bibliography gives the name of the journal in which Falconer (1985) appears as Genetical Research. This is the same journal as Genetics Research; the name was changed about ten years ago.

This guest post is by Radu Zabet on his papers “The effects of transcription factor competition on gene regulation” and “The influence of transcription factor competition on the relationship between occupancy and affinity”

Transcription factors (TFs) find their genomic target sites by a combination of three-dimensional diffusion and one-dimensional translocation on the DNA. We previously developed the stochastic simulation framework GRiP (http://logic.sysbiol.cam.ac.uk/grip/) that allows the realistic representation of the target finding process. The following two papers show our application of GRiP to address a few interesting phenomena:

The effects of transcription factor competition on gene regulation
arXiv:1303.6793

The binding of site-specific TFs to their genomic target sites controls the transcription rate of the target genes. In this manuscript, we discuss the influence of TF abundance on the arrival time of TFs on their target sites as well as the time they stay bound to the DNA. We investigate the TF search process using stochastic simulations and found that molecular crowding on the DNA always leads to longer times required by TF molecules to locate their target sites as well as to lower occupancy. There is also an “emergent property” in cases where many molecules compete in some sort of molecular traffic jam on the DNA. This newly identified noise component may be a contributor to transcriptional noise, by affecting both the size of the fluctuations and the distribution of the arrival times (unimodal or bimodal).

The influence of transcription factor competition on the relationship between occupancy and affinity
arXiv:1303.6869

This manuscript deals with the discrepancy between “predicted occupancy” of a TF to a binding site on the basis of, say, a PWM, in contrast to a “measured occupancy” when we simulate the system with our GRiP framework. Again, we can show that absolute TF abundances play an important role in gene expression, and also provide a compelling case where selecting “the highest peaks” from a ChIP experiment may not necessarily identify the most affine binding sites. Our results showed that for medium and high affinity sites, TF competition does not play a significant role for genomic occupancy except in cases when the abundance of the TF is significantly increased, or when the PWM displays relatively low information content. Nevertheless, for medium and low affinity sites, an increase in TF abundance (for both cognate and non-cognate molecules) leads to an increase in occupancy at several sites.

This guest post is by Matt MacManes on his preprint with Michael Eisen, “Improving transcriptome assembly through error correction of high-throughput sequence reads“, arXived here. This is cross-posted from his blog.

I am writing this blog post in support of a paper that I have just submitted to arXiv: Improving transcriptome assembly through error correction of high-throughput sequence reads. My goal is not to talk about the nuts and bolts of the paper so much as it is to ramble about its motivation and the writing process.

First, a little bit about me, as this is my 1st paper with my postdoctoral advisor, Mike Eisen. In short, I am a evolutionary biologist by training, having done my PhD on the relationship between mating systems and immunogenes in wild rodents. My postdoc work focuses on adaptation to desert life in rodents- I work on Peromyscus rodents in the Southern California deserts, combining field work and genomics. My overarching goals include the ability to operate in multiple domains– genomics, field biology, evolutionary biology to better understand basic questions– the links between genotype and phenotype, adaptation, etc… OK, enough.. on the the paper.

Abstract:

The study of functional genomics–particularly in non-model organisms has been dramatically improved over the last few years by use of transcriptomes and RNAseq. While these studies are potentially extremely powerful, a computationally intensive procedure–the de novo construction of a reference transcriptome must be completed as a prerequisite to further analyses. The accurate reference is critically important as all downstream steps, including estimating transcript abundance are critically dependent on the construction of an accurate reference. Though a substantial amount of research has been done on assembly, only recently have the pre-assembly procedures been studied in detail. Specifically, several stand-alone error correction modules have been reported on, and while they have shown to be effective in reducing errors at the level of sequencing reads, how error correction impacts assembly accuracy is largely unknown. Here, we show via use of a simulated dataset, that applying error correction to sequencing reads has significant positive effects on assembly accuracy, by reducing assembly error by nearly 50%, and therefore should be applied to all datasets.

For the past couple of years, I have had an interest in better understanding the dynamics of de novo transcriptome assembly.. I had mostly selfish/practical reasons for wanting to understand–a large amount of my work depends on getting these assemblies ‘right’.. It was quickly evident that much of the computational research is directed at assembly itself, and very little on the pre- and post-assembly processes.. We know these things are important, but often an understanding of their effects is lacking…

How error correction of sequencing reads affects assembly accuracy has been one of the specific ideas I’ve been interested in thinking about for the past several months. The idea of simulating RNAseq reads, applying various error corrections, then understanding their effects is logical– so much so that I was really surprised that this has not been done before. So off I went..

I wrote this paper over the coarse of a couple of weeks. It is a short and simple paper, and was quite easy to write. Of note, about 75% of the paper was written on the playground in the UC Berkeley University Village, while (loosely) providing supervision for my 2 youngest daughters. How is that for work-life balance!

The read data will be available on Figshare, and I owe thanks to those guys for lifting the upload limit- the read file is 2.6Gb with .bz2 compression, so not huge, but not small either. The winning (AllPathsLG corrected) assembly is there as well.

This type of work is inspired, in a very real sense, by C. Titus Brown, who is quickly becoming to be the go-to guy for understanding the nuts and bolts of genome assembly (and also got tenure based on his klout score HA!). His post and paper on The challenges of mRNAseq analysis is the type of stuff that I aspire to…

Anyway, I’d be really interested in hearing what you all think of the paper, so read, enjoy, comment– and get to error correcting those reads!