After attempting to extract DNA, it’s necessary to run a ‘check’ gel to verify that any product has survived the crushing, spinning, filtering and chemically washing of the plant sample. In a research lab, this involves the use of chemistry a little more carcinogenic than is usually accomodated in public schools. The chemicals used are mutagenic, and the visualization utilizes the kind of UV light that burns the skin if one works around it unprotected. Instead, we used a safe but inexact tool to look at the samples proceed a week prior. tinting is added to the sample and soaked to resolve if any DNA has been found. the initail results were not at all promising

But after a night in the ink, and a rinse in deionized water, I can say with some confidence that we indeed managed to get a DNA sample from the grasses brought from Tampa. Though another verification step away from school will be needed to stand up to the level of inquiry the project demands to move forward, we have passed another milestone in doing authentic research in a high school classroom.

Where at first it seemed I’d found two distinct types of grass, there simply wasn’t a distinguishing characteristic I could find to separate them into different species. I retracted my initial e-mail to my overseas mentor before he confirmed what I had suspected: that I rather had found two variants of the single species Halodule beaudettei. While I was disappointed not to have two species in hand, the improbability of traveling so far to find anything and then being able to key it out to species was thrilling enough. I photo documented the herbarium specimens, and preserved them in a plant press. From the same pressed individuals, I warapped sprigs of grass in seawater-soaked newspapers, along with identifying labels in Ziploc bags. I put the whole batch into an oversized bag, aloing with a sheet of ice I specially poured to fit the sleeve of plants. I wanted to keep the plants as fresh as possible without freezing them into mush. Through TSA security (out with the ice) and on tthe plane from Tampa to Atlanta, then Atlanta to New York, the grass looked almost as green and healthy as it did under southern skies just a few days prior. I speculate that growing in such a muck-y, hot and low-visibility environment, the plant was physiologically hardier than the ones I’d seen in Australia, which oxidized pretty badly in just a few days in some cases. The plant press containing the herbarium voucher specimens eventually got to New York after being mis-routed through Baltimore, but on Tuesday the 29th, everything was in place to extract the DNA from our plants. one hurdle (collection) had been surmounted….another lay ahead.

After getting back to the states, it’s been all but impossible to reflect on the experience of the barcoding work recently undertaken in Australia. The pressure of school and home have left little time to make progress on anything not immediately and practically demanding my attention.

So when a quick trip back to Florida materialized with just a few short day’s notice, I immediately considered the possibility of continuing the seagrass barcoding work begun back in the land of Oz. Most of the time was previously committed, but I managed to negotiate an 80 minute window during a shopping run to jump over a seawall and grab any grass I could find. I parked on a side street, crossed a business park wearing rubber booties and carrying a bucket. I hopped over the seawall and was soon in the familiar mangrove ooze of my home state.

The turbid water did not immediately look promising for seagrass, which is sensitive to reduced light penetration. But after working my way carefully among sprawling mangrove roots, clumps of algae, freestanding oysterbeds and the occasional bonnet head shark, I came to a channnel outlet that led to more open waters. I worked my way a distance along the strand facing the bay before taking a random perpindicular turn into deeper water, seeking groves of grass. Before long, I had some luck, grabbing a handful of long, spindly runners I found. Knowing that such grasses tend to grow in clusters, I charted a widening arc until I found several other such stands. Recalling my work overseas, I was careful to collect large connected segments to yield multiple samples from a single individual. two distinct forms appeared, one long and bolt-y, and the other dense and clumped. I thought I may have chanced upon two species, but had no time to look more closely. It was time to turn back, and hope for the best. the time on the water was wonderful and lonely. Familiar and strange. I was home, applying lessons learned in another hemisphere…using what I’d learned in a new context that was once my home. Standing in the middle of backward, I was heading forward again.

After getting a significant chunk of work done starting the seagrass barcoding effort, late-breaking developments have changed the course of our work to come.  The CBOL, an international body organizing the efforts of individual institutions, published results in the Proceedings of the National Academy of Sciences (PNAS) suggesting the use of a two-loci standard genetic barcode for plants.  The genes they suggest using are the rbcL and matK genes.  This recommendation bears out findings made during our own investigation, but means that we are temporarily on hold pending discussion of this directive among the scientific community.  These are not two genes we worked with (although they were on our list to eventually get to.)  Because the materials consumed can be quite expensive, we will neither proceed with sequencing genes we had planned to check (ACCD, CCS1) nor will we order the primers to check the others until they are finalized.  The indicated time-frame for deliberations is September to October.

Fortunately, the majority of the work completed is the same in any eventuality.  The collected and extracted DNA now in storage at JCU will simply be put through a new PCR reaction, and then sequenced for base order.  Theoretically, this work could be completed in two or three days following final approval of the international standards.  Furthermore, it potentially means that the posting of the results of this work is closer at hand, since at that point they will represent internationally recognized standards, as opposed to the individual effort they have been so far.

I’ve had an intense learning experience here, which has awakened dreams of new possibilities.  But for the moment, the time has come, and I’m on my way to where I am (and have been) needed.  Thanks to everyone whose time and support made this possible, especially M and Em.  I can’t wait to share my experience in person with everyone in the states, soon…..and to renew my companionship with my new friends in Australia, whenever they make their way to my hemisphere.

After hitting a bit of a down note on Tuesday, I spent most of Wednesday and Thursday tidying up affairs in and around the lab.  Unlike my work at the Chang lab in NYC, the space and materials at JCU were shared, and so a careful cost-accounting of my activities had to be undertaken, from the first cell extraction kit to the last  pipette tip.  I also had to leave behind detailed notes of all of the lab protocols I used, and go through various fridges to chuck out what was no longer needed, and preserve those things that would perpetuate my efforts after I’m gone.

But after coming to such a full stop, we found a bit of time to try one last time a couple of the tricky reactions I’d been having trouble with earlier in the week.  Equipment had been serviced and new reagents mixed up, so if there was to be a last chance, it would happen now.  I prepared the master mix, added the DNA from each of my samples, and with hands now trained to the purpose, mixed and spun down the plates on the centrifuge.  We popped them into the thermal cycler, and I programmed in the sequence of steps to yield fresh DNA.  Like a master chef, Kor suggested a bit more magnesium chloride in the starting mix (to help the DNA unravel for copying) and adjusted the temperatures of the stages.  Essentially, add a pinch of salt and let it bake a bit longer.

After running my last load of laundry, I headed back to the lab to pour a gel to be set for our product.  After it was ready, I used a multichannel pipette with twelve tips to load the samples.  It’s like adding a fancy border to many little cakes at once.  I took pride in hearing Paolo, a lab hand famous for his speed and dexterity with the gear say “ooh, fancy pipette work….very nice.”  Coming from him, it was a fine punctuation mark for the evolution of my own skill set.

After 30 minutes, we checked the samples, with the results you see above.  We have results from two distinct sets of DNA collected from the same plants.  One is blurry and unclear, and the other nice and sharp.  This means that one gene managed to copy nicely, while the other, for no lack of effort on my part, simply won’t work according to the recipe in place for it.  It was vindication on two fronts.  On the first, the clean bands can now be processed directly, saving a step in the project after I leave.  These genes should be sequenced by the end of the coming week, and will serve as the third independent barcode experiment I worked on.  The other yet failed, but provided clear enough documentation that the process itself is not working for this gene.  No matter how many times I did it, it simply wasn’t going to work…..so it wasn’t something I was doing to muck it up!

To start anew, different starting strand ingredients will be tried, a simplified thermal cycle will be attempted.  But for the smudge-y gene, we managed to learn something useful in the failure.  So in spite of that it had already been called a successful summer, we managed to wring just a bit more out of my time here.  

The last day reinforced a couple things.  One is the value of persistence and the importance of meticulous documentation.  Especially in lab work, small differences yield important results, and being able to refer back to one’s own trials is essential.  The other is a sense of efficacy.  After seeing what happened, I believe that I know what is needed next, if there was another week or month in the can here.  I know enough to be able to steer the car a little bit, even if someone else is keeping an eye on the map.  Notwithstanding my aching home-sickness for family, I felt a twinge of ‘if only’ before I realized something else:  That whatever work one does in science is purpose-built not to have a neat conclusion.  Every question can and should lead to other, interesting questions.  Although I know where we will go next, I know that the path is a journey with no fixed destination….only mile markers to show the distance traveled from one day to the next, from one discovery to failure to breakthrough, stretching away into the unseen, off in the horizon just beyond.

Chemicals and heat are used to dissolve the gel, which is washed away before the residual DNA is adhered to a thin membrane, called a column.  A separate set of processes then releases the DNA from the membrane into a collecting tube where it is stored.  The normal next step is to see how much DNA is in each sample (quantification) before I can send it for sequencing.  But that’s not what I’ll be doing next!

Besides from working a bit more independently on the gel cleanup, the other challenge of working with the new marker is that it is about half the size of the other two.  This means that it’s harder to isolate, since it moves more quickly through gels and is more diffuse when trying to cut it out a band.  Fortunately, a shorter section of DNA means that less is required to successfully complete the sequencing reaction, but I was nervous as ever about how this run would ultimately turn out.  The gels during clean up yesterday didn’t have the clear banding I saw in the prior two times I’d done this step.   I was tempted to quit while I was ahead, but that wouldn’t be very sporting of me!

Anyhow, after running the final ‘check’ of the product this morning, I learned that whatever DNA was in the gel had been lost in the heavy cleaning process.  Very faint bands yet remained, but it didn’t look like enough for us to process for sequencing.  This means I’m 80% back to the starting point, should I continue seeking to get this one done before I leave.  I’m going to give it the old college try, running the PCR this afternoon, and maybe getting through the gel cleaning, though it’s not likely possible in a half day’s effort.  If I do the clean-up tomorrow, by lunch I might be back to where I thought I was this morning, which means that I’d have to get the results by e-mail.  Certainly not the end of the world, but a bit of a downbeat on this one front.  As Dr. Blair suggested, beginner’s luck never lasts! 

At the same time, I’m also attempting to work out with Kor the correct temperature to run the amplification of the Accd loci, which would give us a fourth piece of barcode information for distinguishing seagrasses.  Since we’re seeking to optimize the process for future work, the relative interest on this piece of the process shifts a bit.  Rather than getting the ‘answer’ per se, we’re trying to find the best process to getting that answer for each species, as a reference for the future.  We’d certainly like the genetic info, but first we need to find the best way of achieving it, going forward.  So alas, it looks like the main body of my accomplishment here is done.  I barcoded the 11 species we looked at for RpoB and RpoC, and I became familiar with the processes (and pitfalls!) of this line of work.  I’ll be returning home much more experienced in my lab skills, particularly with DNA manipulation, and hope to put that to productive use in my employment and teaching!

 After hundreds of steps and five weeks of effort, today we anxiously awaited the results from the sequencing lab, which would be the final arbiter of the success or failure of our efforts so far.  As it was pointed out to me today, there are a number of things one has to do between the last checkpoint (PCR gel) and the final product.  The closest analogy I can think of is making a soup you don’t get to taste before serving it to the guests.  There’s simply a great deal that must be done without knowing if it’s gotten disastrously off track somewhere along the way.  Although the sheer volume of material sent to the unit suggested we’d get something back, I wasn’t especially optimistic about generating any useful data at this point of the project.  As we headed to the Genetics Analysis Facility, I half expected to see a wisp of smoke rising from my 96-well tray and a message reading ‘Surrender Dorothy.’

As I watched the sequencer, I saw several ‘Failed’ sequence reads, but many more which appeared to be going smoothly.  After eighty minutes of individually reading the labels I placed on each small bit of DNA in the sequencing reaction, the machine must use an internal statistical algorithm to indicate which DNA base is at each of the 500 or so positions on the gene I’ve targeted.  This last compilation step alone took several minutes to compute.  Finally,we were able to move the data over to a desktop computer to read with sequence alignment software.

The first file we opened was promising.  There were no egregiously bad segments along the gene sequence, and most of the arcs which represent the machine’s read of each DNA base looked clear and unambiguous.  ”This is a textbook sequence” I was told, but in all likelihood it was one of few such clear sequences in the batch, and that I shouldn’t expect too many more as good as that first one.  Leaving out the few (4) which failed to read altogether, there were over 90 more to go through, for two different genes.  But file after file, we were all impressed and excited at the clarity of the results.  Samples that came from different individuals of the same species bore the same sequence.  Samples from different, dissimilar species were different, more so than  samples from closely related species.  Before I could process what I was seeing, the other researchers were congratulating me on the lab equivalent of a home run in my first time at-bat!  Between strong protocols, close and patient supervision, constant care and a bit of luck, I’ve managed to generate useful sequence data for 11 species of seagrass for two internationally recognized DNA barcode loci.  As of 3:39 am EST, I’m the first person to ever do this.

It may not seem like much, but I’m overwhelmed an extreme sense of gratitude and humility for the small success this accomplishment represents.  My family made many sacrifices to spare me the opportunity to work here, and I wanted whatever I did here to redeem at least part of that gift.  The labs that have hosted me invested time and money to train and equip me.  Other people wanted the chance to be here, too.  Doing my best was the least I felt I could do, and getting such a strong result is an affirmation that all of this was the most it could be.

With two complete, tomorrow I will use my waning time here to attempt to process a third gene through cleanup and sequencing to join those which I have finished.  I’ve got to start learning to manipulate my results in sequence analysis programs, and generate phylogentic trees from the data.  It’s the end of the beginning of this effort, which will continue with or without my help, though I’d very much like to keep involved in whatever way I can going forward.

Of all the days I’ve spent these past few weeks in the lab, today was perhaps the most intense.  There were several factors contributing to the stress level.  First is that I was again undertaking procedures I hadn’t done before as I moved through the various protocols.  Secondly, time is getting short, and there’s no room for error if I want to get my results before returning to NYC.  The materials are super-expensive (1 ml of material I used today costs $800) and as ever, the tasks take a LOT of focus to accomplish.

After getting multiple clean copies of a selected bit of DNA made yesterday, the last step is to prepare the gene for sequencing, which will ultimately tell us the combination and order of bases (those As, Ts, Cs and Gs from your HS bio class) in each species’ version of the gene.  In the chemical processing to get to this point, however, there accumulate around one’s target gene a lot of extra stuff the machine that reads the bases doesn’t like, such as salt, bits of DNA, unused enzymes, etc.  Today I had to get rid of these, and then use flouro-labeled bases to dice up my DNA into pieces the machine can read.  You can actually see the difference in the images above of the pre- and post-cleaning DNA.  The gel on the left is streaky with diffuse, multiple bands, indicating objects of many sizes in the mix.  After, on the right, one can see much less streaking, and nice tight bands of uniformly sized material.

To get to the final image, the plate on the right, I had several more steps to go.  First, I made a dilution of each individual tube, two times, based on my prior day’s result to get the same amount of DNA in each.  I was leveling the playing field.  Next, I added to each sample a chemical that would mark and cut the DNA into lengths with a tag on the end showing what the last base was (**I’ll get back to this bit for the technically inclined.)  Then I had to create a foam, put it into a special tray with little funnels in it, and then run my DNA through it to get all the ‘junk’ out.  Simple as it sounds, it took me from 9:45 in the morning to 6 pm in the evening without pauses.  The work takes incredible focus, since often one is moving 1 critical microliter of sample from one tube to another.  That’s .000001 liters.  A mere capful of soda would contain a couple thousand of them!  So it’s easy to make a mistake, especially when you’re doing each step 100 times over.  My hope is that the sheer number of tries and redundancies in my sample set mean I get a complete result, in spite of myself!  At the end of the day, an index was created (keep track of what sample ended up where was in itself a Rubik’s cube-like task) before it went to the sequencing lab.  I am told that we are first in the queue for Monday, and that we should, with a little luck, have something to celebrate by lunch time!  For all but those interested in the principle that operates the sequencing reaction, see you again, online!

**Extra credit for the technically curious among you:

The reaction that marks and cuts the DNA is interesting.  The target DNA, called a template, is mixed copying enzymes and  the raw materials of DNA, including the individual bases.  Sprinkled among the normal bases in the mix are occasional versions that are special for two reasons.  First, they have a ‘marker’ attached, which can be read by a sequencing machine.  They’re like the others, but they’re special – sort of like the powerball in the lottery, they count for a bit more.  Secondly, besides from being marked and machine readable, they also can arrest the enzymes’ copying of a particular segment of DNA.  So – mixing a few of these into the normal DNA copying process yields lots of little pieces of DNA that got copied complete to a certain point before the labeled base was inserted where a normal base would go.  This event halts the copying of that segment of DNA, leaving a clue we can use to piece the puzzle together.  Putting thousands of such copies arrested at different points through the sequencer, we can tell two things.  One is how long the piece is, and second is what ’special’ base was the last one used which froze it.  With many such bits of info, we can puzzle out the actual combination.  

For example, if we had five fragments with known lengths and end bases:

1. 1 base long, ended in ‘A’

2. 2 bases long, ended in ‘A’

3. 3 bases long, ended in ‘C’

4. 4 bases long, ended in ‘T’

5. 5 bases long, ended in ‘G’

Then we could infer the gene base order is ‘AACTG’….just like an acrostic poem, you read it from top down to get the hidden meaning!  That (in simplified principle) is how the sequencer works.