Friday 7 September 2012

Wednesday 5 September 2012

Digitising X-rays: the digitisation of the Biophysics collection

In this week’s post, I discuss some of the digitisation aspects of the project with special reference to the work of one of our digitisation contractors, MAX (previously MAX COMMUNCATIONS,) who digitised 4,000 of the glass and acetate material.

In October 2011, two of the digitisers from Max Ltd visited the archives in order to familiarise themselves with the collection and carry out some test scanning. The material was quite diverse: photographic prints, x-ray acetates, various sizes of glass plate negatives, folded negative rolls and 35mm mounted slides. The majority of this material required external specialists with the required expertise and equipment to undertake the scanning. A small test file was created composed of quarter plate glass negatives and x-ray acetates and sent to the Wellcome for approval. Both Iain Stringer, who would go on to digitise the collection, and David Cordery, head of Max Ltd, have prior professional experience of working with glass plate and acetate x-ray collections at various institutions  around the UK and we were fully confident of their ability to handle the fragile items and successfully scan them. The test images that were sent to the Wellcome were approved and scanning commenced in December 2011.

 The images were scanned at 300 dpi (dots per inch) at 8-bit (bit rate) RGB (Red Green Blue) using an Epsom V750 Pro scanner. This type of flatbed scanner is reliable and fast and from a preservation perspective, the scanner was suitable for digitising the x-ray acetates as the two- inch gap between the bed of the scanner and the top scanner head meant that it did not press onto the x-rays and so would not cause any further damage to the x-rays afflicted with vinegar syndrome (vinegar syndrome occurs when an acetate degrades and begins to oxidise creating a vinegar smell. The surface often begins to warp and crack and this eventually affects the emulsion. Unfortunately the process is irreversible and it is why digitisation is one of the most effective ways of preserving an accessible copy of the item.)

An example of an X-ray acetate diffraction image from the collection.  The  sleeve caption information has been added to the image. This diffraction image was taken by Wilkins around 1953-1955 and the likely source of the DNA originated from human subjects supplied by Leonard Hamilton and Ralph Barclay of the Sloan Kettering Cancer Center in New York.

I asked Iain to tell me about his experience scanning our material compared to his previous experience with similar collections. He said that the plates, in terms of general condition, were some of the best that he has worked with as hardly any were chipped or broken. The only slight issue that he encountered was that some of the slides were mounted with red strips, the adhesive of which had begun to seep and caused them to attach themselves to their transparent sleeves. In such cases, he therefore had to carefully remove the slide from its sleeve. This required a degree of perseverance, depending on the age and location of the adhesive strips on the slide.  

Regarding the x-ray acetates, I had assumed that this material would be trickier to scan considering the conditions that some of them were in. Iain surprised me by saying that for the purposes of scanning they were quicker to scan than the glass plates. Whilst care had to be taken in handling small, fragile and brittle objects like deteriorating x-ray acetates, the most time consuming element of scanning an x-ray was the post-production.

MAX Ltd provided us with images in three formats:  the raw TIFF original file, the enhanced TIFF amended file and a JPEG file. While enhancing an image can be difficult with regard to obtaining an authentic copy of the original, in a situation where the original is difficult to discern, post-production ‘clean up’ is necessary. The majority of the x-ray acetates retained a degree of visible content and by using Adobe Photoshop post production, it made it easier to enhance the original pattern of the x-ray and compensate for some of the surface damage caused by any deterioration.

Finally, I asked Iain what he thought of the acetate and glass material as a whole. He said:

“ I found the material quite interesting…I’ve learnt more about DNA than I have since school,, good thing about my job that I don’t have to concentrate on one specific thing. X-rays of DNA, diffraction, very interesting. They would definitely make a good print, stretched over a canvas, especially one of the really clear ones like ‘Photo 51’”

I agreed, diffraction patterns such as ‘Photo 51’ are visually striking though I personally am more in awe about the crystalline A-form DNA pictures as there is something rather mesmerising about the symmetry of these patterns. You can judge for yourself however, as these two x-ray patterns are shown below.

A-form DNA

B-form DNA

Sunday 19 August 2012

Will the archives of the future be made of the strand of DNA?

The prospect of combining archives and DNA feels like a plotline of a Twilight Zone episode. What's exciting however is that it is a distinct possibility. The Guardian have just released a story about the DNA inscription of a book ( initially reported in the US journal, Science. The book composed of 53,000 words includes eleven images and a computer program. The 5.27 megabit collection of data created over several days was produced by Professor George Church of Harvard Medical School.

The method they used, in principle, was the same as digital inscription: encoding all the book information into a binary sequence. The DNA base pairs in this case representing 1's and 0's with Arginine (‘A’) and Cytosine (‘C’) representing zero, and Guanine (‘G’) and Tyrosine (‘T’) representing one. The team developed a system in which an inkjet printer embeds short fragments of artificially synthesized DNA onto a glass chip. Each DNA fragment contains a digital address code that denotes location within the original fiDNAle.

What makes DNA such a brilliant medium for storage is its data storage with estimates suggesting a gram of DNA can store 455 billion gigabytes. The data is easily readable and copied and maintains its stability for several thousand years.  

The possibilities are fantastic. To put this in perspective, most digital formats require an upgrade after five years with physical data storage such as DVDs having, at most, a twenty year life span. This is because of the constant change in informational software packages within sturdy digital formats, such as TIFFs and PDFs having 10-15 year maximum life span. A DNA code sequence is therefore more desirable than a digital approximate, but neverless it is an exciting development and will potentially rival the paper record revolution in record keeping. This is an exciting archival perspective.

As a cataloguer and digitiser of the DNA related material of the Kings college London archive such a development is one of a personal joy. It would feel wonderfully apt to have the papers charting the discovery of the structure of DNA are encoded into DNA for future generations.

Tuesday 7 August 2012

August Project Update

The project is entering its final months and a quick update as it what has been happening is in order:

To date, 24,000 images have been produced by our digitisers which roughly breaks down as 4000 glass plate and acetate images and 20,000 images from the paper collection. Over the next two months the remaining part of the paper collection will be scanned. Sections that have already been scanned include papers from Wilkins’ early life, scientific working papers, correspondence with scientific colleagues, papers associated with the history of the research on DNA and sections of his autobiography.

Above, is a low resolution copy of some of the images that are being produced. The example is a postcard received by Maurice Wilkins from Francis Crick dated May 1955 and sent from Paris. The postcard reads: "Having a lovely time telling people about your work and my ideas! Hoping to see you in Cambridge for a quiet weekend - Francis".

Our main tasks over the next few months involve the construction of metadata and copyright and sensitivity checking. The latter is the most time consuming as a detailed survey requires a systematic check of all potentially risky material. Our catalogue descriptions are written at a level to summarize the contents of the physical file but because the images will be accessible individually an item level approach to sensitivity and copyright is needed to be certain that legally and ethically all necessary precautions are taken before publishing on-line. Needless to say this process is time-consuming and has proved to be the most taxing element of the project.

Apart from the construction of metadata and the sensitivity checking the only other main strand of the project to update everyone with is outreach. The project continues to gather interest from its social media sites (like the one I’m writing on now). Besides the blogs, the project has had a presence on Twitter and new images have been added to the project’s Flickr site. In May, the archives participated in a Radio 4 piece on the Wellcome Digital Library which was reported previously on this blog. Alongside this, we were privileged to have been visited by Raymond Gosling in March as part of a television documentary. It was wonderful to meet a contemporary of Wilkins and Franklin and hear from one of the key workers his own experience working at King’s at the time. We are quite fortunate to have a copy of Gosling's original 1954 PhD thesis, titled ''X-ray diffraction studies of Deoxyribose Nucleic Acid' which has been selected to be digitised as part of the King's College London Biophysics collection.