Ten tips to give a great thesis defense

I am going to have my proposal defense on tomorrow morning. This is what I think is important guidance for the proposal defense.

The content is cited from http://www.labspaces.net/blog/1390/Ten_tips_to_give_a_great_thesis_defense


Planning Your Talk

1)      Know Your Audience
Everyone will tell you to know your audience, which couldn’t be truer when you’re planning the introduction to your talk. Sure, there is a big difference between talking to high school students and presenting at a conference, but try to think: who is coming to my talk? If they are all cellular biologists like you, then skip the central dogma slide. But if you have a mix of disciplines you need to be able to explain your work to a biologist, as well as an electrical engineer. Imagine you’re giving the talk to one person with each potential background. Would each person be able to follow it? Sometimes you need to sacrifice some specific details in order to explain the important stuff to everybody. (But you should be able to talk extemporaneously on the specifics if anyone asks!)

2)      Justify Yourself
An introduction is more than just a history of your field up until now. That is, it’s more than a literature review. You need to review the current literature, but more importantly put your research into context. What have you done (or what are you doing) that no one else has done? Keep in mind that just because no one else has done X doesn’t mean doing X is worthwhile- there might be a very good reason why no one else has done it! As you introduce your research you’ll likely explain why you’re doing it, but make sure you also explain why others in the field care. Even more important that justifying your work is justifying your conclusions. You MUST be able to back up any claims with solid references, or solid experimental results! In many cases this means statistical tests of quantitative data. When in doubt, err on the side of “inconclusive” or qualify/temper any of your statements rather than stretch your conclusions.

3)      Tell A Story
One of the most jarring moments in a bad presentation is the lack of transitions. Your presentation should flow from slide to slide and section to section. This will most likely mean that you aren’t going to present your experiments in the order that you did them. You’re NOT telling the story of you working in the lab! Think: what are the overall conclusions from your work and how can you explain and prove the things you’ve concluded? Walk your audience through the story, laying out the evidence convincing them you’re right about your conclusions. One last thing: you’ve (hopefully) done a lot of experiments, you’ve invested a lot of time, energy, and maybe even money into these experiments and you want to show off everything you’ve done. But if an experiment or data slide doesn’t fit in the “story” you might have to leave it out. If you can’t make it fit in the flow of your story and/or you don’t NEED it: leave it out.

4)      Sweat the Small Stuff
The little details are important. Even if you have some really great results to show, you’re going to anger, upset, or at least annoy your audience if you don’t pay attention to details. Some examples:

  • Label the axes of any graphs (with units), don’t use 10E3 mV (when V works) and don’t forget error bars!
  • Make sure any images have scale bars, and label items of interest. (You might know what’s a cell and what’s dust, but everyone else might not!) Use the same size, color, and font text.Try to use the same slide layout.
  • Make all your graphs, diagrams, molecular depictions, etc. with the same program throughout. It’s noticeable if you copied one molecule from a paper, made some in ChemDraw, and others with ChemSketch. The same holds true with graphs in Excel versus Origin.
  • Excel can be your friend but if you use the default graph settings it will be your downfall. Don’t leave on the gridlines or use the standard random colors. Oh, and look into Origin.

5)      Present in Bite Sized Slides
For each slide be sure to explain everything. Explain the x and y axes of your graph, explain what a large value indicates, and a low value indicates. Walk people through how you set up the experiment, how you collected the data, analyzed the results, and talk about the controls. Before moving to the next slide, restate the major finding or “take-away” from this slide. What did this experiment tell you, and what questions are still unanswered. This will help build in transitions as you tell your story. You probably know every piece of your presentation inside and out, but you need to remind your audience of salient points from earlier in the presentation.

Giving the Presentation

6)      Practice, Practice, Practice!
Even the most beautiful slides with the most logical flow and greatest data can trip you up if you don’t know what you’re going to say. It should go without saying that you can’t just read off of your slides, but seriously: practice, practice, practice! Run through it in your head, do in out loud and most importantly, do it in front of other people: schedule practice talks! In the days leading up to your presentation you should be able to run through the talk in your head without notes. As you’re walking the halls, driving, or cooking you should run through the talk over and over. The goal is that when you get up there on the big day, everything comes out naturally- almost second nature. For me, I need to write a script- I don’t memorize it word for word, but the act of writing what I want to say helps. Of course, if you’re a naturally gifted speaker and can give a talk on the fly you’re set- but you should still practice!

7)      Don’t wait until the last minute
The goal of practice talks is to get feedback from friends, lab mates, classmates in general, and hopefully your advisor. It does you little to no good if your practice talks are the day or two right before your talk. You need to give yourself time to integrate their changes into your presentation- both the slides and the talk. I like a formal practice talk the week, and two weeks before the talk. This gives you enough time to change slides, change what you might say, and change the written document (if applicable). If you give yourself enough time you might even be able to squeeze in an extra experiment before the big day to fill any “holes” in your story.

8)      Try out the room and equipment
Not all practice talks are created equal. Sure, you can run through the slides on your laptop in your advisor’s office but you really need to get up in front a group of people- preferably in the same room you’ll be giving your presentation. Not only do you get in the presentation mind set, but you get used to the space, you test the equipment and therefore minimize surprises on presentation day. For example, one talk I went to recently was marred by the screen flashing horizontal bars randomly- it was nearly seizure inducing. Finally, they borrowed someone else laptop but do you really want that stress on your big day? Dress rehearsals are your friend!

9)      Be comfortable with your knowledge
In many cases when you present your research you will be the most knowledgeable person in the room about your topic. Be comfortable with that, and confident that you know what you’re talking about. Professors and especially your thesis committee (whom probably know a decent amount about your topic) can smell fear like sharks find blood in the water. Don’t make it easier for them! Don’t let them know you’re nervous, or might not be sure about something. Confidence goes a long way, BUT don’t let it go too far. Don’t get cocky because nothing is more tantalizing that crushing an OVER confident student. Be confident, but not cocky.

10)   Be humble
You know your research, your techniques, your experiments, and your data. But you might get questions a little removed (or a lot removed) from your research. You might even get questions you don’t know the answer to, or aren’t sure about. The best advice I can give someone going into a defense- even last minute- is don’t be afraid to say “I don’t know.” Guessing, or even worse, making something up, is so much worse that admitting you don’t know the answer to a question. I’ve seen professors who will grill a student and not stop until they say “I don’t know” or they catch them answering wrong (guessing/making something up). You’ll never know everything about everything so don’t be afraid to say “I don’t know”. But it is inexcusable to guess or make up an answer- it will only get more painful from that point on. On the flip side, don’t answer every question with “I don’t know”- it’s not a get out of jail free card!


Old Family Cheesecake Recipe

American desserts are always too sweet, made from pure sugar : )

The following cheesecake recipe is really amazing. Not too sweet, but taste great!

Cite from http://plumberrypie.blogspot.com/2013/01/old-family-cheesecake-recipe.html.

Aunt Helen’s Cheesecake                             

What You Need:

  • 1 prepared graham cracker crust
  • 16 oz. container of sour cream
  • 8 oz. cream cheese
  • 2 eggs
  • granulated sugar
  • vanilla extract

What You Do:

  • Preheat oven to 375.
  • Whip 1 cup sour cream + 8 oz cream cheese.
  • In separate bowl, beat 2 eggs lightly and add ½ cup granulated sugar + ½ t vanilla.  Stir mixture.
  • Add cream cheese/sour cream mixture to eggs and sugar.
  • Stir & blend.
  • Pour into crust.
  • Bake 25 min. @ 375.
  • Cool.
  • Beat 1 cup sour cream, 2 tbsp. granulated sugar, ½ t vanilla.
  • Spread on cool cake.
  • Refrigerate 24 hrs.

Qualified leader in my mind.

I worked as project manager and department director previously. I have experienced confused, existing, awkward, embarrassed, upset, depressed during my career growth. I have to say I am not talented to be a leader. But sometimes leaders need to behave properly at that position. I feel very lucky that I have met one, a really great boss who taught me how to be a leader.

Good leaders are hard to meet but they do exist. Be positive to find yours!

Different leaders have varying concepts about management. Some may focus on the profit of company (sad). Some pay more attention to the benefit of employees (lucky). And some just enjoy the feeling to bully subordinates (tragedy). Leaders have right to be take higher benefit than ordinary employees, they should also take the heavier responsibility.

The qualified leader in my mind should have following characteristic.

  1. Give detailed instruction when subordinates need guidance, help them to improve capability and be confident. Employees are not simply “parts” of the company. They are people, they have feelings. By improving employees’ abilities to promote the whole group is win-win.
  2. Give a clear objective. I hate work without aims. And the leaders always change their ideas are also annoying. The leader should have a clear figure of the project, has budgets and exact plans for expected events and accidents.
  3. Supply Good work/life balance. Work is only part of life. People need take a rest and relax. And then work better. Those leader who are mean and push their subordinates to keep working are nightmares.
  4. Accept success and failure. We can not be successful all the time. The project may be fail sometimes. These might cause the lost of money and reputation. It doesn’t matter. Learn the right way to do is the most important. The leader should accept failure and give their subordinates and themselves chance to improve. Never, Never pass the bunk to subordinates.
  5. Be fair. People who pay more should earn more. People make mistake should get punishment according to the condition.
  6. Do what they said. Some leaders are really sucks. They promise something but never do that. Get rid of them ASAP.
  7. Good at giving control. Give rights to subordinate and make them be charge of things. Meanwhile, kept control of the whole progress. Supervising is also important! Control freak leaders who want to know every detail are the most stupid.
  8. Open-minded. Open to suggestions from subordinate. Be good at find values of employee and help them to get better career. Every employee of a leader would be his/her potential extension in the society.
  9. Have board experience to be able to give advise to subordinate. It’s hard to communicate with boss lack of knowledge of certain project. And they can not give an effect guidance. And worse, they will be the key to postpone the progress of project.

The requirement for computational chemist in pharmaceutical company.

I am always interested in drug design with computational simulation. After going through the limited results from Glassdoor about this, I find the following skills that are essential for the position in a pharmaceutical company (Novartis, J&J, Pfizer).

  1. Written and oral communication skill.
  2. Work with cross-disciplinary teams (Medicinal chemists, Biologists, Computational scientists, Bioassay).
  3. Familiar with drug discovery programs (lead generation, lead optimizatoin, molecular target assessment, HTS triage, library design and evaluation, QSAR analysis, AMDET modeling, ligand-based drug design, pharmacophore modeling, lead hopping, homology modeling, MD/QM simulation, virtual screening, molecular docking, silico modeling).
  4. Software: Schrodinger suite, Accelrys Pipeline Pilot, SpotFire, MOE, ChemAxon, OpenEye.
  5. Linux/Unix (automation script).
  6. Programming.
  7. Strong record of scientific accomplishments.
  8. Management skill.

Payment is around $100,000/y. Work/life balance depends on the position.

Work hard to gain it!

The webpage of Scott Wildman’s Lab has a introduction about major procedures of drug design.