I was looking into background literature today for a class I'm assisting with this coming Spring. The class deals with the mosquito gut microbiome and its relevance to malaria transmission, an issue relevant to hundreds of millions of people worldwide. A paper in the Journal of Medical Entomology caught my eye, mostly because it's another case of "I've never heard of that genus or species before!" The species in question is the Gram-negative Elizabethkingia meningoseptica and these folks found it in the midgut of Anopheles stephensi, a mosquito known to be a vector for the malaria parasite. E. meningoseptica not only appears to have antimicrobial properties but has a toxic effect on malaria parasites as well (in vitro, at least).

This bacterial species isn't newly discovered but it's been known by other names. It's been in the genus Flavobacterium and in Chryseobacterium previously. It's been associated with soft tissue infections, even among immunocompetent individuals (but hey, all kinds of species can become opportunistic under the right conditions).

Here's that mosquito paper:
Ngwa CJ, Glöckner V, Abdelmohsen UR, Scheuermayer M, Fischer R, Hentschel U, Pradel G. 2013. 16S rRNA Gene-Based Identification of Elizabethkingia meningoseptica (Flavobacteriales: Flavobacteriaceae) as a Dominant Midgut Bacterium of the Asian Malaria Vector Anopheles stephensi (Dipteria: Culicidae) With Antimicrobial Activities. J. Med. Entomol. 50:404–414.

Here's another example of the interplay between microbes and phage. The authors call it BREX (for BacteRiophage EXclusion, emphasis mine) and unlike other phage resistance systems, BREX doesn't seem to cleave viral DNA. It doesn't appear to be as consistently conserved as CRISPR, either - its components may primarily be exchanged horizontally (see Fig 6 in the paper) among many bacterial and archaeal genomes.


Goldfarb, T. et al. BREX is a novel phage resistance system widespread in microbial genomes. EMBO J. e201489455 (2014). doi:10.15252/embj.201489455

Parallel emotions

Every so often, I'll crack open Emotional Intelligence 2.0 for some novel viewpoints. The book is set up as a series of lists of strategies for improving social awareness, relationship stability, and the like. In a different context, they'd be called life hacks. The strategies vary in quality and detail but are good starting points. I tend to find them useful in principle but not in their proposed execution.

Here's the example I came across today: a self-awareness strategy described as "Spot Your Emotions in Books, Movies, and Music." The general idea is that you can improve knowledge of your own emotional state and reactions by observing how you respond to art and literature. It sounds straightforward but one comment struck me as questionable.

...when a character from a movie or book sticks in your head, it's probably because important aspects of his or her thoughts or feelings parallel your own.

Is that really true? Do I find characters memorable primarily because I see myself in them or because they are different from me? Is parallel the right word, or would intersect serve a better purpose? How we react to the world around us - including the art and cultural products we're exposed to - can certainly serve as a viewpoint to our own emotions. I'd just like to think that I find characters or thoughts interesting for reasons far beyond their similarity to what I'm already experiencing.

The same philosophy is used to teach reading to children. It could be useful to encourage older students (say, college freshmen) about their own emotional responses to art and literature rather than just their analysis of it. I still think that what they'll tend to find memorable isn't primarily about their own parallel emotions. There's so much else involved that I'd hesitate to use the word probably when discussing anyone's reaction to art.

Another bit of R

R makes it easy to chop up and reassemble data frames, whether it's with the subset() function or with dplyr's filter() function. It isn't always obvious how to do it for a whole data frame. That's where apply() is useful.
This will filter a data frame such that all columns have at least one value less than zero in any row, for example:
newdata.df <- data.df[ ,apply(data.df, MARGIN = 2, function(x) any(x < 1))]

any() can be exhanged for all() to restrict the selection to columns in which all values are less than one.
The same code works for rows: just move the apply(...) in front of the comma, then change MARGIN to 1 instead of 2.

This is an example of how an interesting idea can be a technological success while failing at numerous aspects of execution.

There's often a lot in common between science and art. Practitioners of both fields constantly have to prove their relevance to jaded audiences. Scientists and artists may be equally passionate about the importance of each detail of their work, but too much detail (especially without context!) can derail any relevance. It's a tricky balance. Too much passion and detail is indistinguishable from Time Cube. Lose the detail while keeping the passion and you end up with a TED talk.* Lose the passion but keep the detail and your hard work becomes drab and unpalatable.

I don't mind a bit of pretension but a little goes a long way.

*TED talks have their place. They're essentially sermons, though some are more evangelical than others.

Today's music is Emancipator - Merlion.