TWiV 324 letters

Robin writes:

IBD patients

Have them eat one cup of yogurt (the yogurt, not the cup), and one helping of fresh, well washed (without microbicides), uncooked vegetables or other produce each day.

With regard to the gut microbiome, we’re slowly getting to know our “$元!¥”. Much muckraking ahead. Don’t hold your breath/nose.

The unvaccinated should have a choice. The vaccinated should also be allowed the full benefit of herd immunity by preventing the unvaccinated from diluting that herd immunity by excluding them and their contacts at all public venues. Pictures of the dead and dying unvaccinated should also be widely circulated, alongside those of autistics, to permit comparisons.

On the iPhone & iPad there is a free downloadable Podcast app from Apple.
https://itunes.apple.com/us/app/podcasts/id525463029?mt=8

It also takes non-iTunes podcast feeds in rss and atom formats.

Gertrud writes:

Hi TWiVers,
I just wanted to write a quick follow-up to last week’s discussion about whether post-doctoral experience is necessary in today’s job market. I completed a four and a half year post-doctoral fellowship at the CDC with a long-term basic research career in mind. My transition to patent law three years ago occurred almost serendipitously. While a PhD is strongly recommended in this line of work, post-doctoral experience is not necessary in most cases. However, in retrospect, I have to admit that my time at the CDC resulted in some valuable experience with animal work and immunology, which has proven incredibly useful over the last few years. A patent agent is a jack of all trades, as one is often handed work that is outside of one’s expertise. Therefore, the more diverse one’s bench work experience is, the less time it will take to understand new innovative technologies. That being said, there are some days where I really wish I had a PhD in physics.
Thanks again for all your great podcasts,
Trudy.

Marianne writes:
Hey TWiV team,

I’ve been listening to the debate in the follow up section of TWiV 323 around the value of doing postdoctoral training. I have to agree with both Vincent and Alan with parts of their argument around the situation. I happen to have completed two amazing postdoctoral positions after my PhD, resulting in about 7 years of really gratifying work. I chose not to go the academic route, and now work in a very satisfying job in biotech, that definitely benefited from my postdoc experience. I also spent the majority of my postdoc period working in advocacy around postdoc issues at the national level in Canada, giving me a unique perspective on the situation.

Firstly, I found that personally, my postdoc training was incredibly productive, scientifically stimulating, and really ignited my own personal passion for research. I discovered during this time that I really couldn’t see myself doing anything other than science. This was due to amazing mentorship, and the ability and freedom to follow my research where it led. I also had the benefit of mentors who understood the value of good postdoctoral trainees and treated us like colleagues instead of indentured (and incredibly cheap) “lab rats”.

But through my advocacy work in a national postdoc organization that I helped to found, I discovered that this is not the uniform situation for all postdoctoral trainees. I’ve been witness to a lot of manipulation of the postdocs by PIs. There are many reasons why this can and does happen, but at its core it often relates to the power that a single mentor can have on a postdocs career future, and this is something that can and must change.

There is no simple solution, to be sure. However, I keep coming back to a single, relatively easy to implement solution – recognize that PhDs have a value, that those who earn a PhD are valuable to a lab and compensate them as such. Give them a salary commensurate to their education – and then treat them like junior colleagues. The argument against this has always been that grants ‘can’t afford’ to pay postdocs well and that it will result in fewer postdoc positions- but the data states that we also have far too many postdocs for the PI/scientist positions available. So the numbers say let’s pay fewer postdocs better. Let’s put the effort and training and time into quality rather than quantity. Then, maybe the focus would shift to selecting postdocs who have a valid chance to continue in a research focused career path and save them 4-7 years before introducing the crushing reality that they may not make ‘the cut’ and need to make other career plans. Then the onus then will naturally shift back to PhD programs as the tipping point in career planning; and universities will then need to step up to the plate to show graduate students all of the options for PhDs outside the academic arena. This is slowly beginning to happen as postdocs generally ‘wake up’ to career prospects. But generally far too many postdocs believe they are in the small percentage that will go on to be a PI without the sobering discussion around if their CVs match their ambitions.

I also recognize that there are many postdocs who are talented scientists but would not make great PIs. These scientists should move into newly created ‘staff scientist’ positions, with salaries and benefits that befit the role they play. Again that would give ‘the hands’ that large, well funded labs and PIs need, while providing careers for those with PhDs.

The current situation was founded around an ideal where the postdoc period was short, intense and designed to train you for jobs that were available on the other end. The numbers no longer add up and thus the situation has to change. Does that mean no one should postdoc? No. It means that postdoctoral training should not be a part of the natural progression post PhD. It should be reserved for those who see themselves doing no other job in the world, and are willing to move heaven and earth to get themselves to the career of their choice. And they should have their eyes open to the fact that an academic career is far from a given, but that a postdoc can and does train them for other careers. But to do that, you must select a lab, a project and a mentor very carefully. Today, getting into a postdoc is far too easy and getting out of one is far too difficult. Balance is needed.

As always, I’m always interested in the perspective of your group.

Cheers

Ruth writes:

Hey TWiV team,

Love the podcast. Always have. Unfortunately, I don’t get to listen so much since I became a mother 2.5 years ago but I try to keep up when I can.

Regarding your statement: “Parents who don’t vaccinate their children are idiots”….
In the vast majority of cases, I completely agree. If a child is healthy and their medical history shows no reason why a vaccine might be harmful, then the child must be vaccinated! This is for their own good and for the good of the “herd” (ref: herd immunity).

My daughter is (thankfully) a completely healthy child (as far as we know and touching wood) and so I consider it my duty to her and to all other children out there to have her vaccinated against all childhood diseases.

However, I think there must be room in this mindset to consider those children who are not healthy and who may have some sort of immunological reasons why vaccinations may be harmful to that one child. Or, children who were born extremely prematurely. In such cases, the body of knowledge around how the usual vaccines may affect the child must be smaller and less sure. For example, premature babies are surviving now who may not have done so back when these vaccines were first tested. How do we know that their immune systems are still ok to accept the vaccines compared with babies born at term? Might these be the individuals who should be allowed to benefit from the “herd immunity” created by all others being vaccinated?

I have a friend who’s daughter was born 2 months premature and has survived to be a very healthy 10-month old. Her mother was so very worried regarding the vaccinations as the baby was so tiny at the age she was due to have them. There was no allowance in the conversation with health professionals regarding this prematurity – all my friend wished to do was to delay the vaccines for 1 month (not never do it!). The response was “you must do this now” with no reassurance. Any searching for reassurance online got 2 groups of thought: the “you must do it or your child may die”, or the “all vaccinations will harm your children and the Drs can’t tell us what to do”. No middle ground.

It seems to me that a useful, reassuring project would be for some impartial reviewer to look at what evidence is out there that the usual vaccines are still fine for premature babies up to whatever degree of prematurity (or children with other issues) and collect it into one place (and probably make it more accessible to the layman). Doctors / nurses could direct parents to this information whilst still being strongly pro-vaccine as they should be. Maybe this resource is out there but my friend and I could not find it. She went ahead with the vaccines as mandated by her doctor and her baby girl is as healthy as ever. Great! I just think some more reassurance could have been forth-coming for her.

I expect that children who would have problems with the vaccines are already picked up by the Doctors caring for them and removed from the normal vaccine regimens and given only those which would work properly for them (e.g. inactivated vaccines in those who may develop a dangerous infection from active vaccine viruses). But I don’t know that this is the case…?

I’d love to hear your thoughts.
I will now have to listen to every TWiV until I hear my e-mail and your response. My daughter is going to hear a lot of virology over the next weeks and months!

Take care all of you.
All the best,
Ruth,
PhD in respiratory virology (a long time ago!)
(Nottingham, UK)

Marty writes:

Hi TWIV Team!

Thanks very much for responding to my first note below!

I’ve been continuing my study of mathematical epidemiology and updating my tool kit. My purpose here is to provided an update on my breakthrough in modeling the West Africa Ebola pandemic this weekend! But first here is the weather here in Cincinnati: 44 F, DewPoint 31 F, Humidity 81%, Pressure 30.13 ” Hg falling, Wind 8 SSE with gusts to 17.

At first I was was using an exponential model, then a series of linear models and models were never based on the full data set. These models were based upon total cases (the red line in the plot below). This weekend I had the insight to model cases per day (the brown line in the plot) and everything changed!! I had been reading about mathematical modeling in epidemiology and discovered a paper by Kermach and McKendrick written in May of 1927 – they were reporting on a study they made of plague data in the Island of Bombay in 1905 through 1906. They were studying deaths per week and the data was similar to the Ebola cases per day data. The had derived an equation which was a good fit to their data. Its functional form was

A*(sech(B*t+C))^2

– so simple and elegant!

I decided to do a non-linear regression using this equation on the Ebola case per day data. The result was the green curve in plot. The peak in that curve (on 10/24) represents the day when the number of cases per day started declining (Region D). In this model case per day is analogous to speed and cases is analogous to distance, so total case is just the time integral of cases per day. The blue line which closely tracks the case data (red line) is this integral.

When I saw the fit to the case data it literally blew my mind … its right on for the entire data set!! Using this model I estimated the final number of cases to be about 23.3 K and it will take several more months to get there.

Many refer to Kermach and McKendrick as the fathers of mathematical epidemiology. Think of it this way … they came up with their equation 88 years ago, it works today and they did it without computers!

Mike Walsh writes:

So, it looks like they just fitted the data using essentially a regression with specific trigonometric functions (this is the non-linear component they refer to above) to account for the periodicity of cases (note the periodic rise and fall of the brown line above), while estimating a smooth function of the epidemic peak (green line). Use of these functions to model periodicity is not uncommon in phenomena that display such behavior (e.g. climate research). Regarding the specifics of the email, it is a bit more difficult to follow. For example, it is hard to know exactly what they were doing when they ran into trouble with the “exponential model” described, as it is described without an actual form. Presumably they were referring to an exponential growth function of the form Xt = X0e^λt. But typically we would not use this to make predictions about future numbers of cases, normally we would use an SIR model (or a variation thereof), which is the solution to a differential equation that considers a population as discrete subgroups of (S)usceptible, (I)nfected, and (R)emoved individuals (note: the last group is “removed” from the system by acquired immunity or death), and then describes the rate of change the population experiences moving from susceptible to infected, and from infected to removed. The rate of change, for example, that describes the flow of people out of the susceptible group and into the infected group is a function of the number of susceptible people at time t, the number of infected people at time t, the contact rate in the population (or a subpopulation), and the probability of pathogen transmission given contact. This is the framework that we typically use to estimate the number of infected individuals at some future time point. This is an oversimplified model example, of course. Any particular epidemic scenario would require adjustment and often requires the estimation of additional parameters. Often many additional parameters.

The model estimated by your listener is an excellent effort and not wrong, as such, but it may not be robust to varied inputs, i.e. variation in future data. For example, the model of the cumulative cases (purple line) matches almost exactly the actual cumulative case counts (red line), which suggests that the model has been over-fitted to the data. This is a common problem in the application of regression techniques and requires cross-validation with training and testing datasets. I tried to find the paper that they used for their derived equation, but I couldn’t find it. So, without that and without some more explicit details on how the numbers were used I can’t know exactly what they did, but the gist of it is pretty straightforward, I think, and should follow what I described.

Hope this helps.
Cheers,
Mike

Lance writes:

Hi Vincent & friends,

I was a bit behind with TWiV 316 and I’ve just finished listening. Just a very quick comment about Th17 knockout mice – toward the end of the paper discussion Vincent mentioned that you can get mice which are “Th17 or Th17 receptor” knockout. I think what you actually mean is “IL17” or IL17R” knockout. Though, in fact, this would not be the best way to knock out Th17 cells – if you only knocked out IL17 the cells could still make IL21, IL22, IL9, GM-CSF, etc… The way to knock out the Th17 POPULATION is to knockout ROR gamma t – then the cell population can’t develop and all its effector functions are absent. (Link to the original Cell paper showing this: http://www.sciencedirect.com/science/article/pii/S0092867406011056)

Nevertheless a great episode. Vincent I quite agree with you – this area is really fascinating. Reminds me of the mind boggling paper in TWiV 172 on measles virus and LCMV – where co-infection with both viruses led to recruitment of cells responding to a peripheral viral infection into the CNS and immune mediated damage. Amazing.

BW,

Sebastian writes:

Dear Vincent, Alan, Rich, and Kathy,

We recently learned about TWiV when you mentioned our Kickstarter campaign on episode 315, and our friends were ecstatic. Thanks very much for the shoutout!

We wanted to tell you a bit more about us and why we’ve created miniPCR, in case it’s interesting to some of your listeners. We used to be basic scientists who spent most of our lives in the lab: Zeke looking at C.elegans microRNA genes, and Sebastian dissecting neural circuits. One day we converged on the idea that more people should have access to the tools we used in our everyday work, and experience some of the same excitement. We decided to give it a go and now we spend all our time imagining and making accessible science tools that everyone can use (and afford!).

Our hearts have a special place for science teachers. Most middle and high school labs are still decades away from what goes on in colleges and universities. That’s the only window many kids will have into biology, and to help close this gap we started with miniPCR. We hope to make it easier to learn about DNA in a more interesting way that breaks the big-expensive-black-box paradigm.

We’ve found more support for this project than we had ever imagined. And by working closely with teachers and students we keep discovering new ways in which we can use our science backgrounds to make biology education more engaging.

If you think that you and your listeners would want to learn more and give us some feedback we’d love to be in touch. We love getting the word out and engaging with the crowd.

Hope to hear from you soon,
Sebastian and Zeke

Amplyus – Science for everyone, everywhere

Robin writes:

Fab & Fc

From a slash & burn type retired quasi-science person:

Fc = fragment crystallisable (fragment “constant”) – the stem of the “Y”

Fab= Fragment antigen-binding – the limbs of the “Y”

For those who think solar panels are a long term solution (or that alt-fuels will run the Disney World and Wal-Mart lifestyle:

FYI

Compiled by
Robert Callaghan

xxxxxxxxxxxxxxxxxxxxxxxxxxxxx

http://motherboard.vice.com/en_ca/blog/there-are-no-substitutes-for-the-metals-in-your-smartphone

xxxxxxxxxxxxxxxxxxxxxxxxxxxxx

http://theenergycollective.com/barrybrook/471651/catch-22-energy-storage

xxxxxxxxxxxxxxxxxxxxxxxxxxxxx

“Plundering the Planet” – Club of Rome: Shortage of Key Resoruces Ahead – Kontext TV:

http://youtu.be/7jFFGYwOQX4