Tuesday, July 12, 2011

Double-sex butterfly

This article was in the Guardian today, talking about a bilateral gynandromorph butterfly which emerged at the NHM's Sensational Butterflies exhibit last week. The individual is a Great Mormon, Papilio memnon, a polymorphic mimetic swallowtail (like my favourite Papilio dardanus) from South East Asia. The genetic control of the polymorphism by a supergene was demonstrated by Cyril Clarke and Philip Sheppard in the 1960's.

Photograph: Kevin Webb/Natural History Museum, via guardian.co.uk
I have talked about this before. This specimen is a bilateral gynandromorph - the two sides of its body are composed of cells of different sexes. Here, the left hand side is male, showing the characteristic male pattern of this species and the right hand side is female. The genitalia are likewise mismatched - one male clasper and one half of the female reproductive system.

This was how it looked this afternoon; rather tatty. When this butterfly dies (likely in the next week or so) we will examine it much more closely to examine this rare phenomenon. Watch this space!

Sunday, June 19, 2011

What do we do all day? Part II

If you go to the Natural History Museum and take the Darwin Centre tour, you will pass through part of the now famous cocoon. Part of this looks out into some of the museum's molecular labs, and you can watch scientists (like me) at work. This is the view we get when we look back (at you!) from our labs. But what are we doing in there with robots, labcoats and pipettes? Why does the Natural History Museum have a molecular lab at all?

One of the things I have been working on is using the museum's collections of insects (housed in the lower floors of the cocoon) as a source of DNA for molecular studies into the evolution of mimetic wing patterns in butterflies. I take legs from specimens collected in Africa (some over 90 years old) and preserved in envelopes or pinned in drawers.

I can get all the DNA I need from a single leg, even for these very old specimens.

The butterfly legs are soaked overnight in a special buffer solution along with a proteinase enzyme which helps break down the tissue and allows the DNA to diffuse out but doesn't destroy the specimen. After this, the leg can be dried and returned to the specimen and the liquid can be purified to extract the DNA:
This is the filter I use to purify the DNA. When liquid is passed through, the white filter layer binds DNA whilst the rest of the buffer comes through the bottom. After washing, I can wash the DNA out of the filter with an elution buffer and I have clean DNA from a 90 year old dried butterfly. This DNA can then be used in a PCR reaction to selectively amplify the genes I am interested in:

A PCR machine
PCR product imaged using a dye which binds DNA and makes it visible under UV light

The PCR product can be sequenced (you probably didn't know that the museum has its own in-house DNA sequencing facility).
I can now use this in my analyses, just as I would do with DNA sequences from fresh specimens. Such work is only possible at the museum because of the priceless collections and the facilities in the state-of-the-art Darwin Centre. I hope that if you come to the Natural History Museum, you do take the time to go on the Dawin Centre cocoon tour and learn more about the cutting-edge research taking place there.

Wednesday, June 8, 2011

Going collecting in a collection

Museum collections are a truly unique source of data. Some important recent work has demonstrated the catastrophic impact the fungal disease Batrachochytrium has had on amphibian populations The conclusion that Batrachochytrium appearance was coincident with dramatic population decline was made possible by the ability to examine specimens from decades ago, preserved in museum collections.

A single field trip can only take a snapshot of a population, and laboratory culture of specimens necessarily removes organisms from their natural environment. Museum collections typically consist of a number of specimens collected by different people at different times and from different place. In one single room, we can have an overview of the total diversity of a population, a species, even entire orders of organisms through both time and space.

Traditionally, this diversity was analysed in terms of morphological data, principally the presence or absence of certain characters, and was studied to shed light on the relationships between different groups, the adaptation of species to their environment and the changes in appearance of organisms over time and space (when you have specimens from across the range of a species in a single drawer, the differences are much easier to see).

Now, museum collections are increasingly seen as a source of molecular data too. One part of my PhD involves extracting DNA from preserved museum specimens of butterflies, some collected over 90 years ago. These specimens offer me the chance to sample the entire range of one species in one go (the alternative would be a MAJOR field trip covering most of sub-Saharan Africa). It also includes members of populations which are now extinct - without the Natural History Museum collections, it would be impossible to study these populations and I would not have as much power to resolve the questions I am researching. Techniques for making use of preserved material are really only in their infancy: as we gain more experience and as DNA sequencing technology becomes ever more sophisticated, the amount of information we can get from previously collected (and studied) material will be pushed ever higher.

Saturday, June 4, 2011

Geneious Part II

I'm writing a short introduction to Geneious for use in our lab and I thought I could put a bit of it here:

Importing files
Sequences can be imported to a folder using File>Import. The Geneious autodetect feature is pretty accurate and supports .ab1 files from our sequencer, or nexus/phylip/fasta if you have previously editied sequences. To import a Sequencher project, you have to export the contigs from Sequencher in CAF format and Geneious will import these as contig assemblies with chromatograms.

Dealing with chromatograms
Geneious can assemble contigs of the forward and reverse traces. Simply select all the files you want to assemble, click 'Assembly' and choose either 'Assemble to reference' (e.g. for assembling primers to a sequence) or 'Assemble by name' (for matching forward and reverse sequences) and choose if you want Geneious to trim you sequences or not (I leave this ticked as it works really well).

Geneious will give you a report of which chromatogram traces assembled and which didn't. You can elect to save this list by checking the box next to 'Save assembly report' in the previous step.

You can now edit the contigs.
The piece with the red underline is the piece Geneious has auto-trimmed due to low quality. The chromatogram is still there for this piece, but it isn't included in the consensus sequence (at the top) and won't be included when we export the sequence for alignment. If you want the chromatograms to be bigger, you can click and drag the peaks upwards, or to make it take up more of the screen, change the number next to 'chromatogram' in the Graphs menu. A word of warning: the default colours of the peaks are different to Sequencher, but can be changed if you like by clicking 'edit' next to the colours dropdown box on the right:

When you make a change in the chromatogram view, Geneious puts a coloured bar under the changed base. Hovering the mouse over this tells you what change you made and when. This is a brilliant feature, very useful when we are editing our alignments. When you have finished, save and close the editing window to return to Geneious. 

When you have edited your sequences, select them all and hit 'Alignment.'

Choose 'create alignment of consensus sequences' to use your edited sequences, ignoring the trimmed bits. You now have an alignment file, which you can open and check to remove primers etc.

If you want to check a sequence, click the little arrow to the left of the sequence name and Geneious takes you straight to the contig (in the main window). As before,your editing history for the sequences in the alignment is visible as bars under the sequence. When done, you can re-align the sequences by selecting the alignment, hitting 'Alignment' on the top bar and selecting the method you want to use (Clustal, Geneious's own algorithm, the MAFFT plugin etc.). To install plugins, go to Tools>Plugins.

To build a tree from your alignment, select it and hit 'Tree' in the top bar. 
Plugins are available for MrBayes, PhyML and PAUP*, provided you have the executable for these and now the path to them. The full choice of options, model selections, partitions etc. can be typed in. The trees can be opened and viewed in Geneious.

To export any file (sequence, alignment, tree) File>Export and select a file format (nexus, fasta, phylip, genbank flat etc.).

Wednesday, June 1, 2011

What is a species?

'Species' is a familiar term, familiar both to biologists and the wider public. But what, exactly, is a species? How do we define them? What does it mean to say that scientists have found a new species?

To a large extent, this is a philosophical argument; there are a multitude of definitions of species (referred to as 'species concepts'). To some degree these reflect the different reasons people have for classifying organisms: chef vs taxonomist vs geneticist for example. Some species concepts are abstract and theoretical, others purely pragmatic.

Most students of biology will be familiar with the Biological Species Concept - that individuals belong to a species if they can interbreed to produce fertile offspring. This is a reasonable starting point, but is more or less impossible to use or test. For paleontologists looking at two fossils, or forestry workers wanting to know if a particular insect is a dangerous pest or harmless, it is not going to be possible to try and engineer a mating! Similarly, we cannot test a new discovery using the biological species concept. A further problem is it is not always (or indeed often in the case of plants) in agreement with what we think of as species. Species hybrids (the offspring of a mating between parents of different species) exist, and can be fertile: with similar hybrids, with either parental or even sufficiently closely related species.

An example of a hybrid 'species' Papilio nandina, a cross between P. dardanus and P. phorcas

Where does this leave us? Certainly if we are interested in purely pragmatic identification, variants of Ecological Species Concepts are attractive. These all start from the assumption that all members of a species occupy the same ecological niche: they live in the same kind of habitat, have the same lifestyle and eat the same things. This is great if we want to tell edible from poisonous or pest from harmless, but these ideas are not without their problems. For starters, in many organisms (such as butterflies) different life stages have different lifestyles - they occupy different niches:

P. dardanus caterpillars

P. dardaus adult feeing on nectar
If we wish to learn about the biology of such organisms, it is going to be more than a little bit inconvenient to have different life stages in different species! This (perhaps unfair) criticism of ecological species concepts is in addition to other practical difficulties - primarily that it is very difficult to determine an organism's niche (or to see if two similar organism really are in the same niche) without extensive study. For a bucket full of beetles in alcohol or a drawer of pinned butterflies, we simply don't have the data to make a purely ecological idea of species workable.

Some scientists have taken advantage of technological advances to suggest new concepts, such as using particular pieces of DNA as a molecular barcode: if a specimen has the barcode belonging to a species, then it is from that species! There is now an initiative to generate a barcode from as many species as possbile - some advocates of this approach promise Star-Trek like machines which can use DNA barcodes to identify any biological material which is fed into it.

Whilst we obviously all want to own a tricorder, I'm afraid I don't think this is an end to the problem of species. We might want what we define as a species of bird today to be comparable with a species of dinosaur, for instance.Personally, I think that if we want our species to be biologically meaningful, we need to think more broadly, to be (dare I say it) a bit Post-Modern. What we call species (and I'm tying my flag firmly to the mast of the 'species are real' camp here) are best described and circumscribed by using a range of metrics, reflecting the vast range of reasons people might want to identify a species and the range of data available for different taxa.

Monday, May 30, 2011


I hope you caught the first episode of the 2011 BBC Springwatch season. Whilst watching, it struck me that the film and photography featured is going to be a fantastic resource for biologists to come. As scientists become ever more aware of the importance of context and as we wish to study behaviour and the dynamic properties of organisms, the action of live organisms is increasingly a point of study. When thinking like this, video like that on Springwatch is equivalent to the data preserved in museum collections (containing preserved individuals and associated information): data which we use to understand the species and groups the specimens or film represents. These recordings are specimens, just like anything in a museum cabinet.

All collections need curation and the BBC (in addition to all of the education, programme-making etc.) curates a HUGE collection of wildlife film, certainly no small task! Worth the license fee? Absolutely!