Can being a writer actually make you a less efficient scientist? For the past few months I have been knee-deep in a high-throughput RNAi screen for pathways that affect cell shape and the actin cytoskeleton. Automated image analysis has come a long way recently, thanks to the work of people like Anne Carpenter, whose CellProfiler program revolutionized the field. But detecting the textures and shapes of subtle morphological characteristics against the backdrop of normal cell variation is still in its infancy, and nothing so far can beat the discriminating power of the human eye. Although we have a few talented computer scientist collaborators working on the question, I’ve still had to sift through nearly fifty gigabytes worth of images by eye.
Worth a thousand words: But only a dozen are permitted
Not only do I have to look at these cells, but I have to describe what I see using a controlled vocabulary. From this descriptive annotation, we can boil everything down to numbers, and thereby start to cluster and compare all the various manifestations of our gene knockdowns. Biology to numbers, and numbers back to biology.
So far, so good. You would think, as a writer, that describing what I see would be the easy part. My bioinformaticist collaborator assigned a list of about ten features, aspects such as “cell shape”, “actin”, and “cell number”, and told me to tick “yes” if what I saw differed from the negative control cells, or “no” if it didn’t. There was also a Notes field at the bottom where I could jot down any additional comments, which the bioinformaticist said was usually left blank.
But soon, as I began sifting through the images, the parade of weird and striking manifestations quite overwhelmed the simple digital system. Just in the “cell shape” category alone, the word “yes” seemed insufficient to cover the reoccurring and surprisingly regular morphological themes that unfolded: triangles; star-shapes; bipolarized elongated spindles; rhomboids. Ruffled lamellae pushing in multiple directions like a fractal pattern. Sharp-edged vertices with precise actin-rich points at their tips.
Surely, I reasoned, if these shapes were well represented enough to become familiar to me, it would be a shame not to record the information; and this was true for all the other categories too. If I did it in a way that could be retrieved later, surely we could subdivide the original descriptors in retrospect and make our clusters that much more informative. The bioinformaticist gave me the green light, so I began to use the Notes section. Copiously. Some of the images were so complicated that I ended up with paragraphs. I was incredibly careful, or so I thought, to use consistent words and phrases and to subdivide these with consistent punctuation for ease of automated text-mining extraction later.
Of course it all went horribly wrong.
“You’ve spelt ‘multinucleate’ twenty different ways,” the bioinformaticist informed me dourly. “I’d hate to be your editor.”
Spelling was the least of our problems. When he tried to extract phrases separated by commas, it soon became clear that I’d mistyped or missed out commas in many places, or used them in list series as you would in prose. When he decided to bin the phrases and just take individual words (1274 words with more than two letters, to be exact), the decoupling of adjectives from nouns – and the mingling of different aspects in the same paragraph – was deadly. Of course it’s so clear now I should have stopped annotating early on as soon as I realized the scope of the problem, subdivided all the digital descriptors and started over, but at the time, it had all seemed workable. So I’m cleaning up the list now, and will need to go back and manually reannotate. The bioinformaticist is trying to restrict me to twenty terms; I am secretly fomenting rebellion.
When scanning down the spreadsheet of all the nouns, adjectives and adverbs extracted from the Notes section, removed from context, I feel heartily embarrassed. Did I really see fit, over those long weeks, to use terms such as “beads-on-string”, “curviness”, “cross-hatched”, “raggedy”, “stellate”, “splotchy”, or (oh God) “fried-egg”? My inner novelist, impossible to avoid. But it did make me start to think how Henry Gee the great writers of the ages might have annotated my screen. Perhaps Homer would have seen a “wine-dark” quality in particular nucleoli, or considered a wonky spindle to be “well-greavéd”. Robert Browning, when contemplating a wispy actin phenotype, could have evoked the long grass amongst Roman ruins where lovers lie, “such a carpet as, this summer-time, o’erspreads/And embeds”.
And I like to think that P.G. Wodehouse might have deemed a particularly spectacular set of stress fibers “dashed decent”.
I suppose “Ooh, shiny!” might not be too descriptive?
Heh. We could file “ooh” into the adverb column and consolidate it into the “positive modifier” camp.
“Shiny” might be de-duped into the “increased expression” adverb set, with a bit of luck.
This really is no fun.
But it did make me start to think how
Henry Geethe great writers of the ages might have annotated my screen.Words … um … er… fail me.
Henry, if you come anywhere near my adjective list, all is lost.
I don’t know where I could file words like “chthonic”, let alone most of the terms found here.
So I have a few cloves of garlic here, and I’m not afraid to deploy them.
Oh, I think ‘irksome’ describes HeLa cells quite well, actually.
How about creating a tag cloud of your text? Not because it’ll helps (it might, who knows?), but because we deserve to see how important “splotchy” is in your vocabulary.
Mandragorous. Pustulocrustaceous. Hamandrokrystic. Verkorean. Trinistaguline.
There. I feel better now.
Bob, that’s a great idea. But I think I might die of humiliation in the process. I can’t even look at the spreadsheet without blushing.
Richard, are you dissing my beautiful HeLa cells? I might be forced to react like a provoked mother bear.
They are beautiful. In much the same way that a supernova is beautiful.
My DPhil supervisor used to refer to them as the ‘anti-cell’.
Damn. I just missed the perfect opportunity to say “pulchritudinous”.
The good thing about them is that they are large and flat and don’t move around. Very sexy actin filaments, beautiful protrusions. And I’ve got them completely domesticated; there are practically making the tea.
Also a fantastic set of publicly available microarrays. If you believe in that sort of thing.
Damn. I just missed the perfect opportunity to say “pulchritudinous”.
Worry ye not, Richard – there’ll be another one along in a minute.
Jenny raises a serious and very general problem in science – how to turn qualities (multifactorial, continuous, analog measures) into quantities (discrete, digital measures).
This is more of a problem in
propertraditional biological disciplines, such as phylogenetics and systematics, than in molecular biology, because the substrates of these disciplines – natural objects – must be classified in a way which, inevitably, seeks to reduce a continuous distribtuion into a set of discrete intervals. Battles rage in the thickets of cladistic analysis about how to divide a character into sets of ones and zeroes.And describing shapes is extremely hard except in terms of analogy. Morphologists are beginning to harness the awesome power of computer graphics and digital animation (including such techniques as motion capture and rotoscoping) to quantify complex shapes, and, more, how these shapes change with time: everything from how a bird’s wing moves, to how the shape of an embryo changes as it develops. But this is largely a recent phenomenon.
Back when I was doing my PhD, trying to tell the difference between the fossil bones of aurochs and bison, a difference that turned up consistently in the literature, and which I was able to verify in my own samples, concerned the shape of the metacarpal (the front cannon bone). In the aurochs, the diaphysis of the bone flared smoothly outwards as one moved down the length of the bone, towards the epiphyses. In bison, by contrast, there was a kink, as if the bone had ‘shoulders’. The best way I could describe it was by analogy. Aurochs cannon bones had the smooth taper of burgundy bottles, whereas bison bones had the sharply defined shoulders of claret bottles. Everyone I talked to immediately grasped the concept.
Jennifer – you should have been a Drosophila geneticist, then you could have given free reign to your creativity in naming all those genes – though they didn’t too bad without you, with dishevelled, frizzled and the like.
I’ve heard talk of providing controlled vocabulary and ontology support to authors – so you can do semantic markup as you write your paper. But what you are highlighting is a need for your own controlled vocabulary-building tool.
I guess this problem has been around for a while – things like Endnote let you put in your own keywords on papers and these can easily get out of control.
Lastly, it’s ironic that this post on controlled vocabulary has no tags …
Homer was blind, at least.
(The thousands of words that have been spilt over that “wine dark” sea to which you refer.)
The adjectine “ubiquitous” was usually applied to actin in my day, and we left it (sniffily, no doubt) at that, as myosin was more glamorous and tantalizing.
I like your post very much; and it makes me realise that creating a hierarchical keyword index for the whole of Nature Publishing Group content from 4 November 1869 to present day is a doddle by comparison with what you are trying to do.
Where’s D’arcy Thompson when you need him?
Henry, you are very right. Reducing biology into numbers is an amazing challenge. But I like to remember my digital camera: from zeros and ones I can shoot videos that look and sound almost real. Burgundy and claret bottles – now those are descriptors I can embrace.
Frank, a common set of descriptors would be challenging: every sort of cell has its own quirks. In fact, it was starting with the lab’s preexisting terms for cultured fly cells that led me to underestimate the complexity of the human cells.
Tags! (she scurries off to make amends)
Hi Maxine, our comments crossed.
Is Nature really trying to do that? I hope you have some excellent text-miners at your disposal. And when they’re finished maybe they can come visit me!
Well, it could just turn out to be a thought experiment. Time will tell! I love thinking about things like this but they are always so ambitious.
Maxine, now that I’ve witnessed first-hand how easy it is to toggle actin levels, I feel that calling it a ‘housekeeping gene’ and using it for equal loading controls in experiments is a terrible idea.
Yes, myosin is much more glamorous. But I’m a simple soul.
Actin & myosin loading controls are the work of the devil himself. I decry, malign and fail to anoint any blot with such nonsense.
So there :p
I just looked at Henry’s tag cloud… funnily enough the first one on the list of “Falling” tags is “Biology”…
Biology. Hah! Who needs it?
Those rising and falling tags are very strange, when I looked at mine for some
hideous torturegame of M@’s. I think they are really “most recent” (rising) and “older” (falling) rather than frequency-related. That’s my theory anyway.Jenny, on the actin, I agree it has turned out to be a much more interesting molecule than many people thought in those days. It was in the Eocene, or threabouts, if that is any cause for your forgiveness.
Re Frank: “things like Endnote let you put in your own keywords on papers and these can easily get out of control.”
I have only just realized that my little problem is not just restricted to screen annotation. As Frank notices, it’s going to apply to anything where you need to describe. And descriptions of anything remotely interesting just don’t lend themselves to brevity. How many years would Nature Network folk have to tag things, for example, before the “phenotypic clustering” would start to become useful? Or just look at my “network” here; it’s become too big to be terribly helpful.
How can we limit out descriptors without losing resolution? Is the key in subdivision, like Russian dolls? Or do we learn to live with a poverty of description?
Hierarchies? Although can one apply them retrospectively, when we see which of our “subcommunities” (eg our personal network here) has grown and could do with the odd subdivision, compared with those that are still sparsely populated? I have no idea but Euan Adie and co will know I am sure.
This reminds me of the difficulties associated with qualitative coding in social science research. For example, from a series of free-form discussion groups, how can one conclude what the prevailing attitudes are towards HPV testing when there is so much variation in the way people express their concerns and beliefs?
Having not used this technique myself, I only know that it’s difficult to do well. Perhaps someone familiar with this might be able to help you set up a coding scheme?
Interesting, Hilary. Will check out your links.
Maxine, Matt Brown told me there was no way to apply a further hierarchy to your network beyond friends and contacts, and no plans in future to allow it.
How about “arch enemies” as a useful category?
How about “arch enemies” as a useful category?
Note to self: Don’t drink coffee while reading NN.
“But detecting the textures and shapes of subtle morphological characteristics against the backdrop of normal cell variation is still in its infancy, and nothing so far can beat the discriminating power of the human eye.”
I just gave a talk yesterday about something that could not be scaled up to high throughput RNAi screen level, and only worked if you scrutinized pictures on confocal. I’m trying to write a manuscript as/after I write it all up for my thesis, because only the screens that work ever get published, and I think the balance is off. Developing a screen for changes at the (sub)cellular level is not as easy as it seems, and more of the “this didn’t work” stuff needs to be out there.
Eva, I know people doing high-throughput high-content confocal timelapse. Just an SP5, but it seemed to work quite well.
But I agree it would be nice to know more about what doesn’t work.
Jennifer and Eva–Nature Precedings will happily take manuscripts/posters/presentations describing “what didn’t work”, as long as it is described in sufficient detail to be useful for other scientists. We’ve only gotten a couple of manuscripts along these lines, possibly because researchers haven’t thought to use preprint servers in this way. In addition to helping other scientists, one might get a citation or two from those that find the paper useful (which is probably better than having your hard work sit in a drawer somewhere). That being said, we’re also happy to take papers describing things that “did work”.
Re: Categories on NN, if you’re introducing “arch enemy” then what about “frenemy”? Somehow I suspect some research involves creating a lot of frenemships…
My goodness. I shall have to publicize that fact. Thanks Hilary (my entire scientific adventure, somedays, feels like “What didn’t work next”).
Hilary – I know about Nature Precedings! It’s one of the outlets I’m keeping in mind. BMC also has something, but with different guidelines.
I am repeatedly telling people that these places exist….
Jennifer – well, it turns out that the things I was knocking down as controls were quite resistant to knockdown. It looks like they really need all their protein gone to do anything, so it works on a small scale, when you can keep an eye on that, but once it’s scaled up you kind of expect that most things are being knocked down properly and/or that a partial reduction in protein would at least show some change in phenotype, but that didn’t seem to be the case for my system of interest.
“What didn’t work” journals could certainly benefit from very detailed tags, so you could find what didn’t work quickly and effectively.
hmm…
“Everything” might be a broad church.
heh. Subdivided into ‘everything you actually tried’, ‘everything you were too afraid to try’ and ‘everything you were going to try until you lost heart’.
Yesterday, the comment I aborted because of its uselessness was “sounds like fun”.
My two cents today is that the same problem of controlled vocabulary applies to nosology. Thus we ended up finding mutations in a gene which can cause a clinical entity known respectively as Matthew-Wood syndrome, PMD syndrome, syndromic microphthalmia (MCOPS9) or atypical PAGOD or Fryns syndrome, depending who you ask.
One person’s aplasia may be another’s agenesis. That’s why it’s easier to work with the same clinicians repeatedly, who tend to use a restricted and fairly reproducible vocabulary. Although as you’ve seen, it may not be all that restricted.
The chemists have the right idea: IUPAC rules the chemical nomenclature with an iron hand, and woe betide anyone who violates the conventions. And it’s not static: the committee meets annually to track the trends and amend anything that needs tweaking.
“One person’s aplasia may be another’s agenesis.”
Rarely a day passes without the same thought crossing my mind.
Oh, my dear Charles. Your wit is wasted in this humble salon, sir. You should be on what is known as the Stand-Up Circuit. I have heard that the fashion of throwing Rotten Vegetables is very much on the wane, an evolution which can only work in your favor.
Going back to NN subdivisions, I (like Barbara Axt in her profile) don’t “get” the difference between friends and contacts and am still trying to puzzle it out. Arch enemies, now — I suppose the editors on the NN would be very easily identifiable 😉
Richard, do you mean that if you spill your coffee over your keyboard while on NN, you are your own worst arch-enemy?
Maxine, I prefer to think of myself as an arch-nemesis.
Has a certain something about it, don’t you think? A touch of class.
And here I thought you were a particularly intelligent arch-anemone.
Perhaps for the categories, “contacts” can remain one-sided to constitute a network, whereas “friends” should have mutual approval. I’ll let you all develop on that.
How about “secret crushes” and “unrequited lovers”? According to the Guardian, NN is just an elaborate scientific dating service.
We wants a link, please?
Why, Heather: are you ‘looking for someone’?
GSoH, own grant, enjoys long knockouts and RT-PCR
[attempts to salvage the tone of this comment-thread and so restore a ladies’ honor] I’ve looked at the picture at the top of this post and it occurs to me that it might be possible to classify such pictures according to their resemblance to the abstract works of various surrealist artists, such as Roberto Matta
or Yves Tanguy
although probably not Paul Delvaux.
Here is it, Heather; check out Matt Brown’s patented smoky gaze.
Henry, re: “attempts to salvage the tone of this comment-thread and so restore a ladies’ honor”
You’ve picked a bad time to make an apostrophe/singular/plural error – it looks as if Grant is awake.
That’s a lovely Tanguy. I wonder what gene you have to knock down to get that?
Yeah, but lady’s doesn’t look right.
If you drop your gel on the floor it’d be ‘doing a Motherwell‘
and if it were Pollock‘ed
you might as well pack up and go home.
And people wonder why the funding bodies are chucking so much dosh at ‘sci-art’.
I can tell you how you get that Tanguy – you puncture the yolk sac with your forceps. I did that this afternoon.
Jennifer was more on my wavelength than Richard – who must be dozing now. All apostrophers are safe for the moment. That article must have caused a wee bit of consternation in the NN ranks, no? Why does everything have to get back to Facebook in particular?
And I am amused that 39 in England is not considered “early stage” and for some reason, in France, it is. You grow up suddenly at 40 here.
Jennifer was more on my wavelength than Richard – who must be dozing now. All apostrophers are safe for the moment
Apostrophe man never sleeps. Your apostrophes will be cared for at any time, day or night.
Heather, I was a bit peeved they took my comment out of context and exposéd my age at the same time. And don’t worry – being a postdoc at 39 is practically terminal.
(mentor cap on)
You probably don’t need more validation in your career choices – but to hell with programs and passively following an apparently typical channel to your goal. I can come up easily with ten personal acquaintances, all examples of highly successful scientists, who did not get to the research position in which they found that success until their 40’s. And the success itself – it’s all a matter of definition, isn’t it?
Comments out of context: that’s most of the general press for you. I’d highly recommend, to anyone who’s made it this far in the thread, the excellent A Scientist’s Guide to Talking With the Media by Richard Hayes and Daniel Grossman.
(takes mentor cap off and toddles off to bed)
That article is just plain odd. Another case of Feynman’s bongos – the desperate need of the media to find some ‘human interest’ angle for any story involving scientists because otherwise ‘real people’ won’t be able to connect with them.
(there is a famous picture of Richard Feynman playing bongos – the story is he got rather upset about the fact that whenever he was mentioned anywhere, including physics textbooks, it was seen as necessary to use that photo rather than one of him doing, say, physics)
I second Heather’s recommendation. My copy has so many pages folded over at the corner that I should probably start marking the important stuff by unfolding them.
Heather, I am wondering if those ten people stand out in your mind because they succeeded, and how many other failed.
The SMC’s Soundbyte guides are good too.
Heather> My guess about everything being linked to Facebook is because it is percieved that “everyone knows what a community is with gadgets and networking”. A few years back they tried to say “my space” but now Facebook is that thing…. or was…
I found this “…the scientists who use it are mainly in their late 20s and early 30s and may be hoping to discover a life-, as well as a research-, partner online” to be fun. I am guessing the key word here is “may be hoping” since it doesn’t really say that we/’they’ are actually but that we ‘could’. So it is not a true lie and of course they stuffed researchpartner as well as love in the same sentence. You can’t say it isn’t clever?!
So, never mistrust a “media” person to use your words to best suit them. Not to be evil, mind you, just to make a living… thanks for the link to Heynes&Grossman – left mine buried in a box in an attic across the Sea.
and I should have used a better sentence than “never mistrust..” ah well, that happens late afternoons. hopefully it’s understandable.
Take heart, thirtysomethings. I am convinced that education is wasted on the young. And I’d reprise a famous anecdote, as follows. The scene is an ecumenical conference on advances in human reproduction very topical – Ed. The discussion spills out into the bar. At a table, a minister, a priest and a rabbi discuss the issues.
“Life begins at conception,” argues the priest.
“Life begins at birth,” the minister retorts.
This futile cycle continues to revolve for a while until the priest and the minister realize that the rabbi hasn’t said anything. He’s just been sitting there drinking his beer.
So they ask him: “Rabbi? What’s your view?”
“About what? When life begins?”
“Sure.”
“That’s simple,” says the rabbi. “Life begins when the kids leave home and the dogs are dead.”
Update:
The bioinformaticist has come up with a quick and easy way for me to consolidate all my terms (including the 20 different spellings of ‘multinucleate’) and we’ve introduced a hierarchy system that should be able to map everything back to sanity.
Fingers crossed!
details?
This sounds like openwetwarewiki stuff. But I’m too lazy to look up the links (and they’re going into a blog post of my own, anyway).
Nothing that exciting, alas. I need to visit every one of the 1200+ words and assign it to one set of controlled words (about 30 at the moment). Then each of these 30 will be assigned one of the original digital categories. I’ll do this in Excel. Then he’ll run a simple script to realign the entire dataset accordingly.
As an example, all instances of words describing multiple nuclei will be assigned one term, “multinucleate”, which will be associated under the category “DNA”.
We don’t see any automated way of doing the first step, because only I know what I meant by the various words. But I don’t think it will take very long.
The only thing we lose is degree – the adverbs. But I don’t think it matters so much if a characteristic was X vs. “very X” or “not very X”. The main thing is that X was perturbed.
I will also have to make a special pass to fix key descriptions that I originally used to modify more than one noun. The word “disorganized”, for example, was used to describe both actin and microtubules. I will need to revisit the raw data and manually associate the two in every case.
What a job. Phew! Manual reassignment of every one of 1200+ words to 30 categories? Just goes to show that the last step of even the most complicated analytic procedure always boils down to a human judgement of qualitative value. Any statistical procedure will give you a probability value – but it’s always up to you to lick your finger, stick it out the window, and judge whether, in your opinion, it means anything.
The only thing we lose is degree – the adverbs
“Adverbs are overrated,” said Hermione, waspishly. “What we want are more adjectives. More juicy, plangent, pungent, lambent, refulgent, grandificent, terebinthic, markoretic adjectives.”
“Lambent” – I think I might have used that one!
Re: The Grauniad piece… “He predicts that scientists who post comments, blogs and data from experiments on sites like Nature Network will eventually be allowed to count these as part of their research output.”
In my recent blog post I mentioned an ex-editor (odd descriptor…) I spoke to recently who gave me some career advice. He said I should be writing a lot, and meant non-techncial writing. I mentioned Lablit and Nature Network and he was impressed. Hopefully the slightly downer tone of that peice isn’t representative of the “world” at large…
Ian, that’s interesting. I can’t see blogs like mine ever being considered as part of research output, but those sorts who are fond of writing a “News and Views” style blog about a particular finding, and applying some clever synthesis, might well apply. Still, where is the quality control? Anyone can spew out 1000 words, but if I’m competing for a job with a prolific blogger, would it be be fair to be judged by that metric? I wonder.
I would be embarrassed to count my comments and blog posts as part of my research output. Really. It’s part of my personal output, the way that volunteering for a patient association or climbing peaks >3000m can be. Could go on the CV, but not really as impressive as the other occupations in my opinion.
Writing anything is indeed good practice if you want or need to write for your career, as most of us do to some extent. It doesn’t have to make it out in front of the public, though, to be valuable in that respect.
I agree, Heather. Much as 99% of what we do in the lab doesn’t make it into the public eye either!
And if it does it needs washing out with sterile saline.
>rimshot
I was thinking more about what Hermione-via-Henry was saying about adverbs. Actually, I think whether a quality increase or decreases after perturbation may actually not matter so much – the degree and direction may not add extra information. We’ve probably all had experimental systems in which you express a little bit of protein and something happens, but you express too much and the opposite happens. The system is responding to something in the works, via perturbation, but it could respond either way. I think if, say, actin is upregulated via perturbation, the agent you are knocking down could be part of the pathway but not necessarily a negative regulator. I need to think more about this idea but I think it might have some substance.
So…maybe my lost adverbs won’t matter too much in the grand scheme of things.
Jennifer, I agree with you in principle but have reservations about throwing the baby out with the bathwater. After all, the whole principle of microarray analysis is to find sort-of quantitative differences in gene expression. These would be expressed in English by relatively, a little under-expressed or dramatically up-regulated.
Couldn’t you keep the adverbs sort of integrated into the quality of the adjectives you do keep? As in “filiform” versus “elongated” to denote a sort of gradation in how stretched out something is, at least to yourself?
No, you are right. I don’t believe a single pixel of information is dispensable. What I am going for now is a compromise between detail and expediency. And the original
novelNotes section will always be associated with each gene; I just won’t be able to use every pixel of information for the purposes of clustering the data.Jennifer> I guess you can save those 20 some words that you ended up with and then create a file for the next time you look at the cells and then just tick them off in order?! “multinuclear, yes; elongate, no; beautiful, Yes…” and then get a nice matrix with different cells and reactions!? I guess that is one good point with your ‘sorting through’ study now? To make it easier if you want to redo it?! And of course interpret. Imagine, you can combine different parameters and intpret if things are (looks) linked or not 🙂 oh, it’s fun to do when the file system is made proper and you can just write a little program to look through it…
Actually, Asa, the reason we are doing this is so that we can cluster genes with like phenotypes together. The theory is that genes in the same pathway are likely to give similar phenotypes when knocked down (yes, this is a gross oversimplification, but with some pathways this does turn out to be true). So if we can transform our 20-odd categories to numbers, and then compare all the permutations, we can generate a tree and see which genes behave similarly when knocked down. From this, we can see where unknown genes cluster (are they behaving like a set of known genes?). And then this is a shortcut – we can guess where to start characterizing them biologically.
But you are right, the list will be used again the next time, and we can compare data sets accordingly.
It will be fun, once I’ve cleaned everything up!
Jennifer> It sounds fun and interesting!(Although, I know from another ‘similar’ study of looking for those trees and changes but not in cells that it might turn out to be square eyes and numbers rolling before your eyes… but as with everything, there are more fun parts than others!)
It does help a million though to have a good set up and defined words etc 🙂 I look forward hearing some more aout the words in the end – how elaborate did they turn out to be!?
It’s not as bad as we feared. A lot of it really was spelling and synonyms, which is ease to map back to a sole word. As for the rest…well, I just need to roll up my sleeves and do it the old fashioned way.