08/27/13

Lodgepole pining

Another summer is drawing to a close and another school year starting, meaning that a few of us have wrapped up some successful field seasons!  I spent a good chunk of my summer travelling across BC and the Yukon collecting lodgepole pine needles for DNA samples in my project looking at the genetic basis of local adaptation to climate, a ground-truth of the main AdapTree project. I’ve returned with  2,585 trees sampled from 122 different provenances grown across 16 different test sites I visited, as set up by the BC Ministry of Forests in the Illingworth provenance trial as well as a smaller provenance trial set up in the Yukon Territory.

I just want to share a few photos and fun stories to give a feel of what my field season was like, but first I have to acknowledge a lot of very great people. I had a lot of help along the way, and these are the people to whom I owe many thanks!  First off, Sally, Ian, and Kristin, head of and in the Aitken lab, for all the help given in terms of advice for the field, getting gear together, and letting me borrow gear, supplies, and a great 4WD field truck. Second, to Nick Ukrainetz and Vicky Berger with the BC Ministry of Forests for their immense amount of help providing me detailed information on all of the provenance trial sites I visited, teaching me good safety practices for driving on logging roads, and providing me with a VHF transceiver. Third, a huge nod to my advisor Mike Whitlock for all of his support in making these collecting trips possible and imparting knowledge over the phone of what to do when we were being swarmed and chased for >20km by what we thought were angry bees in the woods. Fourth, to Anne Berland, graduate student at the University of Victoria for joining us in the field, not only being great company, but also an excellent navigator to finding our sites and to letting me borrow a second  pruning pole from their lab. As well as to Stilianos Louca, UBC grad student, who without having been in a car accident in the Yukon (no one hurt) I would not have been able to visit test sites in the Yukon for sampling and received his help in exchange for helping on the long drive back to Vancouver. And lastly and also most greatly to my two amazing field assistants for my main sampling trip: Evan Cronmiller and Warren Neuvonen, UBC undergraduates, who endured hot days, cold nights, hard ground, tall trees, biting insects, scratching branches, swarming flies, long drives, slingshot fatigue, and in the face of it all remained happy, friendly, energetic, hard-working, and continued to keep me laughing and smiling on our whole trip.  I really could not have done it without all of these people and have the good fortune of being able to now spend my fall semester extracting DNA from all of my samples.

So, on to the fun stuff.  What is it like to sample lodgepole pine for 3 weeks in the woods?

First off, you see lots of trees, and lots of logging roads.

 You drive around a lot, and often it is hard to find a site (the sites were established in 1974, so just under 40 years old!), but sometimes it is actually quite easy!

And you encounter many logging trucks along the way.

The loggers would often talk to us on the radio, curious about what we were doing in our truck way out in the middle of nowhere.

Short tree from a provenance in Yellowstone

The trees can be very tall, or they may be very short.  We clearly could see effects of local adaptation going on in the field.

Holding a 40-year-old tree.

We used slingshots to sample the unreachable branches on the tallest trees. By the end of the trip, it was no problem to shoot a tree and get a branch to fall in one shot. The slingshots got so much use that we had to buy replacement straps halfway through the trip. But we sometimes resorted to extreme measures to sample when trees were too sparse for the slingshot.

Warren and Evan became pros.


These included throwing the pole to reach a just-out-of-reach branch or climbing a tree.

We had a lot of really, really buggy days where the incessant buzz in your ears almost makes you go mad. 

But you manage to find your peaceful and pretty moments.

And enjoy some amazing campsites, like this one on the Bluewater River in the Canadian Rockies.

We often got covered in pollen, walking out of the woods with yellow boots and pants.

And I spent a lot of time each evening filling my samples with silica gel to dry them out for the long trip back to the lab.

And in the end, I even was able to see the Yukon and sample there!

Stilian with the pruning pole

So thanks again to everyone for all of the help and making this a productive and memorable summer. Stay tuned in the future for the results of my project!

08/16/13

Conference organizing tips

By Sally Aitken and Jack Woods

We are recovering from being heavily involved in the organizing and scientific committees of Forest Genetics 2013, an international conference that brought together ~180 people from four forest genetics organizations and 22 countries. This is the third time that Sally has played a lead role involved in organizing a conference (and we both hope that it’s our last). We enjoyed the meeting despite all the work, and want to share some lessons from these experiences for those of you who may be taking on a similar task for the first time.

1. Location, location, location – if you want a good turnout, pick a location that people want to go to. We picked Whistler, BC as in the summer this ski resort village has ample accommodation, a huge range of summer activities, and is in a mountain environment. As it is a village and not a city, there was lots of opportunity for impromptu interactions among delegates outside of the conference schedule, with people running into each other as they strolled through the village or enjoyed a beer on an outdoor patio.

lost lake

Lost Lake – within walking distance of the meeting venue | Photo by Susannah Tysor

2. Give people free time or they will make it – if you fill every spare minute of every day, participation in talks and poster sessions will drop off as delegates bail to go sightseeing or take advantage of available activities. The better the location, the more important this becomes. We scheduled a free afternoon midway through the conference for people to do some hiking, cycling, relaxing, or to go up the gondola, As a result, we had a full audience for most presentations. If you have to run more concurrent sessions to make this work, then do so.

Graduate students climbing

Graduate students Simon Nadeau, Ian Maclachlan, and Joane Elleouet enjoying their afternoon off | Photo by Kim Gilbert

3. Networking doesn’t happen during talks – Make sure that you keep sessions running on time (with brutally efficient timekeepers to keep those speakers under control who believe what they have to say is more important than the next speaker or break), and schedule enough time for people to chat at coffee breaks and meals. Otherwise, these conversations either won’t happen, or will happen in the hallway during presentations.

4. Select invited speakers carefully and save some heavy hitters until the end – Too often, conferences end with some poor graduate student speaking in a concurrent session to a group of empathetic friends squirming in a near-empty room because most participants have left before the end of the conference. You can avoid this by keeping the conference relatively short, and by wrapping up the meeting with a session of well-known invited speakers. Develop your invited speaker list with your scientific committee, and be sure to include a wide diversity of speakers in terms of age (and career stage), gender, scientific topics, etc., rather than inviting only the usual suspects.

5. Poster sessions need care and refreshments – It might seem easy to skip a poster session if you aren’t presenting, but these sessions are important to plan as people put a lot of work into their presentations and often are grad students who benefit greatly from feedback on their projects and new contacts. Adequate time needs to be allocated to these, and a free drink and snacks often provide the motivation for people to attend. Try to keep the posters available as long as possible, as most people end up chatting during the poster session and will enjoy the opportunity to sneak in at some time during the conference to quietly view the posters (or, for our introverted colleagues, to get away from the crowd).

6. Food matters – The quality of food seems to stay in many people’s minds after meetings are over, so get a foodie or two to pick items from the available menu, and don’t skimp on amounts. Make sure that you give people registering the option to state dietary requirements or preferences, and pass these on to the caterers. Vegetarians are common these days, but (being vegetarians ourselves) we find that you need to provide vegetarian choices to omnivores as well, as otherwise meat-free options disappear quickly, leaving the vegetarians with only dead animals to eat. If possible, have a different caterer and venue for the banquet than for the lunches for variety. A common problem with coffee breaks is that there is a traffic jam because there are insufficient or poorly designed stations with food or drink, and this can delay the resumption of sessions (as it did for us on the first day). Talk about this with the caterers, and insist on an adequate number of stations. Make changes if needed as the conference progresses.

7. Budgeting and payment– Budget to break even with the minimum number of people you expect to register, not the number you hope for. Develop a budget with “must have” and “nice to have” items (meals, SWAG, entertainment, free drinks, etc.). You can add items from the “nice to have” list as registration numbers climb. Include both high and low estimates for all costs and for numbers of participants as you estimate registration fees. A little “gaming” with those variables will help you determine your likely break-even point. Then budget conservatively from there (i.e., set registration 10% higher than you think it needs to be). Developing a budget spreadsheet with cost and revenue estimates with high and low estimates for attendance and costs, and that allows some easy gaming of different scenarios helps with estimates. PayPal is useful for payments, but it comes with a cost that needs to be budgeted for. You can also hire a conference management company to handle payments and many other details, and make your life easier, but for a small meeting, this will increase your registration fees substantially.

8. Conference SWAG – None of us need more cheap plastic mugs or ugly, poor quality conference bags. However, think about items that could decrease, rather than increase, the environmental footprint of the meeting. If the caterer uses disposable coffee cups or provides bottled water, then give people reusable cups or mugs and provide tap water instead. We were pleased to find a source of conference bags made of recycled street banners by a non-profit society called Common Thread employing women with various types of barriers to finding work. A portion of the cost of the bags also went to the Whistler public art fund. The bags were beautiful, unusual, and a big hit.

9. There will always be hiccups – Develop a vision and a plan with your conference committee through consensus, and then do your best to realize those plans, but accept that there will always be a few things that go wrong, and a few people that will be unhappy about something. Don’t fret about these, and don’t change your plans for one or a few highly opinionated people. Similar to planning a wedding, it’s pretty much impossible to keep everyone happy, so just do the best you can and stick to your plan and vision. I was given this advice early on by a friend and colleague (thanks, Jean Bousquet!), and used it frequently.

10. Don’t overcommit – Think twice before you plan to speak at the conference you are organizing. You will be busy with conference details and preparing a talk may be the final straw. Instead, try to highlight your lab’s work through profiling your postdocs’ or grad students’ presentations.

11. Don’t bend to all specific requests – some delegates will seek special provisions, refunds at the last moment, want travel arrangements changed for field trips, and generally drive you crazy with requests. If the arrangements are clear, don’t be too flexible or you’ll be caught in a swirl of last minute details that can be difficult to keep track of. State the rules for refunds, deadlines, etc., clearly on the conference website and stick to them. Delegates are adults and can deal with their own issues.

12. Be cautious with fraudulent registration attempts – there are at least two types of fraud to be careful with. One is where people will attempt to get invitation letters to use to acquire entry visas to the country. They may register and pay, and seek the invitation letter under the guise that they need it to get travel approval. The second may be where people pay with a stolen credit card and then seed a refund. There are likely others. If something seems odd, look into it before you send a letter or refund.

 

08/15/13

SNP calling

Identifying single nucleotide polymorphisms (SNPs) is an important step in the AdapTree project. Currently, we are getting back exome sequence data from ~1300 individuals in two species, interior spruce and lodgepole pine. We need to accurately identify variant sites so we can use this information to determine which sites are potentially responsible for climatic adaptation (e.g. SNPs associated with phenotypes and climatic variables). We are also using these SNPs to develop a 50k SNP array in both species. SNP quality is important, as we do not want to waste space on our array with falsely identified SNPs. On the other hand we also do not want to miss important genes that might be under selection because our SNP calling criteria are too strict.

I have tested various SNP calling methods using exome re-sequencing data from 12 interior spruce samples. I tried Bowtie2, BWA (mem), Picard (mark duplicates) and GATK for indel realignment and base quality recalibration. For SNP calling I used mpileup with and without BAQ as well as the Unified Genotyper from GATK. My results are in a series of blog posts on the Rieseberg lab blog and I hope you find them useful. Please let me know if you have any suggestions for SNP calling. We only want to do the alignments and SNP calling once for the entire set of samples, because it is going to take a long time!

SNP calling I – alignment programs and PCR duplicates

SNP calling II – Creating a reference for GATK and Picard

SNP calling III – The indel problem

SNP calling IV – Base quality score recalibration

SNP calling V – SNP and genotype calling

SNP calling VI – Variant filtering