Just found this clip on Youtube of a small performance I did in Irvine last year:
Just found this clip on Youtube of a small performance I did in Irvine last year:
I recently played through Valve’s Portal. Even though the game is approximately 10 years old, the graphics still hold up rather well. The biggest thing that surprised me is how far integrated graphics chips have come in the last decade; I remember when I was first played this game, it required a very good graphics card for its time. Now my laptop can run it rather smoothly!
Sorry for the hiatus – back to our regularly scheduled programming!
In this post, we transition from the “classical” methods of organic chemistry, and move to modern material. The “classical” reactions are those generally taught in undergraduate organic chemistry, and while reactions such as oxymercuration, alkynylation with acetylide anions, and PCC oxidation are no doubt useful, they are not used that much anymore. Reactions dealing with mercury and superstoichiometric amounts of chromium are no longer palatable in today’s environmentally conscious era.
One of the holdovers from classical organic chemistry is the necessity of conducting reactions with as little water as possible. Water is generally thought of as a “bad” solvent, one that will rapidly quench any reactive species present and bring everything to a grinding halt. This thought process is not unfounded; after all, when working in the lab, frequently you will quench a reaction with water before working it up in order to extract any products formed. However, given the recent interest in Green chemistry from the chemical research and manufacturing sector, there is now a lot of interest in developing water-tolerant reactions. These reactions also have the added benefit of being milder, but the caveat is that one has to put more thought into extracting and purifying the organic material afterwards.
The papers covered in today’s post are on Shu Kobayashi’s work on water-tolerant Lewis Acids and their application in organic synthesis. This paper really marks a distinct gap between “classical” and modern organic chemistry, because when most people think of Lewis Acids, they will think of Friedel-Crafts promoters such as AlCl3, FeCl3, Al2Br6, BrF3, and others. These are very strong Lewis Acids and are also notoriously water-sensitive; they all react with water or undergo hydrolysis. One of Kobayashi’s early papers from 1998 demonstrated the possibility of doing a Lewis Acid-catalyzed Mukaiyama aldol reaction with water-tolerant Lewis acids – this is a big step from the previous versions of the Mukaiyama aldol reaction, which commonly used TiCl4 as the Lewis acid. Even this Evans’ asymmetric aldol reaction makes use of some very water sensitive reagents – namely, n-butyllithium and dibutylboron triflate.
Kobayashi’s main insight was that both the hydrolysis constant and water exchange rate constant (WERC) were critical features for determining if a metal salt would be a good Lewis acid in aqueous media. Basically, if you can choose a metal cation that has a low enough affinity for water (as determined by the hydrolysis constant), but yet can exchange it’s ligands with water at a fast enough rate, you have a good aqueous Lewis acid. This can be seen from the figure below – all the lanthanide cations are good Lewis acids because they have WERC values and hydrolysis constants right in that sweet spot. It’s like Goldilocks – not too low, not too high.
This simple observation then opens the door to a whole plethora of possibilities. The next question is – are asymmetric reactions possible in aqueous media? The answer is… yes.
The ligand 8 in the figure above is a chiral bis-pyridino-18-crown-6 derivative, but the point is yes, asymmetric reactions are possible in aqueous media! I mean, this should be no surprise – all biochemistry is asymmetric, and it occurs in aqueous media too.
Friedel-Crafts reactions are also possible with these lanthanide triflates in aqueous media, but the issue here is reactivity. A traditional Friedel-Crafts reaction with benzene generates a benzenium ion as the intermediate, which will immediately quench itself with any adventitious water present. Therefore, one can only do aqueous Friedel-Crafts reactions involving less reactive (or more reactive depending on how you look at it) species, such as indoles.
So now you’re familiar with one of the most important advances of modern organic chemistry – water-tolerant Lewis Acids!
I’ve been meaning to write something about Theranos for a while, and seeing this rather dramatic article in Vanity Fair yesterday spurred me to action.
Theranos is a Silicon Valley company that was started by Elizabeth Holmes as a 19-year-old undergraduate student at Stanford. I don’t have any personal involvement or interest in the company, but the story of Theranos is reflective of the biotech industry as a whole, and as a rather large company with a multibillion-dollar valuation, all eyes are on it as well as the other large startups. Back in 2014, when it was around the time of my graduation, people were telling me to look up Theranos and apply there since “it was hot” and “Liz Holmes was going to change the world”. In hindsight, I’m extremely glad that I dodged that bullet.
I remember reading this New Yorker article shortly afterwards and feeling a great deal of skepticism. One of the things that tipped me off was this passage:
“One day, in her freshman year, Robertson said, she came to his office to ask if she could work in his lab with the Ph.D. students. He hesitated, but she persisted and he gave in.
[…] That summer, at the Genome Institute, Holmes worked on testing for severe acute respiratory syndrome, or SARS, an often fatal virus that had broken out in China. Testing was done in the traditional manner, by collecting blood samples with syringes and mucus with nasal swabs. These methods could detect who was infected, but a separate system was needed to dispense medication, and still another system to monitor results. Holmes questioned the approach. At Stanford, she had been exploring what has become known as lab-on-a-chip technology, which allows multiple measurements to be taken from tiny amounts of liquid on a single microchip.”
Over the years, I have worked with many high-school students and undergraduates in research, and I also did “research” myself in an organic synthesis lab while in high school. The one thing in common that all new researchers have is this: they don’t know anything. I’m not saying this to be mean, but to lay the reality – working in a research lab is a vastly different experience from doing coursework. For instance, all undergraduates will study diazotization of anilines and learn the variety of transformations they can undergo (Sandmeyer and other reactions). But carrying out one of these reactions yourself is vastly different from simply drawing the structures and reaction arrows on paper; you wouldn’t know just how explosive diazonium salts can be unless you have actually worked with them before. One of the key things you learn in research is humility, which is why getting any kind of research degree is often described as an ego-shattering process; 99.99% of the time, when you think of something, chances are, it has been done before.
Here’s an interesting story that illustrates my skepticism: There was a high-school student who used to work in my lab when I was doing a PhD. We all knew that this student had no interest in science, beyond getting some “research lab experience” to bolster her CV and improve her chances for admission to an Ivy league university. She would come and “work” for only 2-3 hours once a week every Friday. Anyone who has any kind of experience doing research or any kind of lab work knows that you can’t get anything done on that schedule. This student had never set up a single reaction from start to finish (which involves setting up the reaction, monitoring it, quenching it when complete, working it up, purifying the crude, isolating and weighing the product(s), and finally characterizing the product(s)). And yet somehow she managed to win first place in the state science fair, presenting a chemistry project with practically no self-generated data!
That’s why I’m skeptical about “child prodigies” in science, because it takes a long time to develop the foundational knowledge required to make serious contributions, or even to understand the subject matter properly. I’m highly doubtful that after doing basic “research” for a few months, one would have the necessary domain expertise to be able to start a company. I’ve been studying chemistry for 12 years and I feel like I don’t have the necessary expertise! To put things in perspective, one of the criticisms about Theranos is that “finger-stick blood tests aren’t reliable for clinical diagnostic tests; because the blood isn’t drawn from a vein, the sample can be contaminated by lanced capillaries or damaged tissue“. This is true, and anyone with a proper understanding of high-school biology would be able to tell you that. Another issue is statistical – when your sample sizes are smaller, your error bars are going to be correspondingly larger, and this is an important consideration when you’re trying to do measurements on vanishingly small concentrations of analytes (oftentimes ng/L). I guess this would be an instance of people succumbing to groupthink. I mean, the premise of Theranos is awesome, don’t get me wrong. Miniaturizing diagnostics is a huge challenge, and is on the cutting edge of science, engineering, and medical research. George Whitesides (Harvard) is actively working in this area, as are many others. But is it really possible that a 19-year old could solve a problem that the smartest people in the world are struggling with? Color me skeptical.
I remember I was once watching the lectures from Stanford’s Intro to Chemical Engineering class a few months ago, and I stopped watching in disgust once I realized that the instructor, Channing Robertson, was now on the board of Theranos.
Also, I remember my father asking me multiple times about how Theranos was able to secure so much funding if the scientific foundation was so shaky. This article explained everything:
“[…] none of the big V.C. outlets invested in Theranos. When the company raised an additional $200 million in early 2014—which gave Theranos a $9 billion valuation and made Holmes “the world’s youngest self-made billionaire,” worth about $4.5 billion (on paper, a point that few stories ever noted)—that money largely came from private equity.
You couldn’t find Michael Moritz, John Doerr, or Peter Thiel on the Theranos board. And while Marc Andreessen has repeatedly come to Holmes’s defense—blocking Twitter followers who have questioned her and even implying that she could be the next Steve Jobs—his firm, Andreessen Horowitz, did not invest in Theranos. (And even those V.C.s who did are now trying to distance themselves. Theranos is no longer listed among Draper Fisher Jurvetson’s “featured investments,” even though its logo was there this time last year.) When I’ve asked V.C.s why they didn’t pour millions of dollars into a company that appeared to be changing the world, I was told that it wasn’t for lack of trying on Holmes’s part. She met with most top venture firms. But when the V.C.s asked how the technology worked, I was told, Holmes replied that it was too secret to share, even to investors. When they asked if it had been peer-reviewed, she insisted once again it was too secret to share—even to other scientists.”
But that Vanity Fair article was eye-opening. I didn’t know that Theranos’ chief scientist ended up committing suicide due to the pressure and unreasonable expectations put upon him. Yikes. That scenario can be traced back to Holmes’ lack of scientific training; as I mentioned before, a proper experience in scientific research and a proper scientific education will teach you humility, as well as the fact that the laws of nature bend for no one.
I just saw yesterday that another paper has been published based on work I did in my PhD. I was aware of this because the student wrapping this project up was in touch with me, writing up the paper from one of my thesis chapters and requesting copies of the characterization data (NMR, GC-MS, and HRMS) for all the compounds I had synthesized. This paper is the continuation of the Organic Letters paper that was published several months ago, and I’m relieved that I finally have a first-author paper, even if it is only in Journal of Fluorine Chemistry (which has an impact factor of 2.2), and coming a bit too late (over a year and a half after getting my PhD!) to be useful. But hey – it’s another line under the “publications” section in my CV, and at this point, anything helps.
In July-August of this year, The Economist featured a series of 6 articles on seminal ideas in economics. These should be required reading for everyone, not just students of economics, as these concepts underlie so many macro- and microeconomic decisions that affect our lives on a daily basis.
I’m glad that The Economist did this, as it is important for the general public to know these theoretical concepts and how they originated (that being said, the readership for this magazine does skew towards the more educated populace).
There are a lot more topics that can be covered in future variations of this series, such as “Supply-side economics”, Behavioral economics (including Kahneman’s work on Prospect theory), the Ricardian model of free trade and how it breaks down, and modern work on globalization, among other things.
I just finished the immersive Data Science bootcamp by Logit on Friday and am still slowly recuperating from the experience. The course was intense – it was a firehose of material, and a rapid survey of 2 years worth of material at a Master’s level compressed into 12 weeks.
The course started with a tour of vanilla Python and the Data Science related packages (Numpy, Scipy, Pandas, Matplotlib, Seaborn, Scikit-Learn, statstools, and many more), and then covered basic statistics and probability, before going into Machine Learning, which was the main part of the course. Both unsupervised and supervised models were covered, as well as the major methods of regression, classification, and clustering (e.g. K-Means, K-Nearest Neighbors, SVM, Naive Bayes, Decision Trees and Random Forest). Regularization, resampling, and feature selection were also covered, as well as transformation (e.g. PCA). After making a simple midterm project to do an analysis of a publicly available data set (I chose to work with the Boston Housing Dataset), we moved on to Neural Networks, Time-series analysis, Natural Language Processing, and “Big Data”. As usual, if you are curious about the course materials, I’ll be uploading some of the assignments on my github.
As I mentioned above, this course was really tough. The fact that I was also commuting 2 hours each way did not make it any easier, either. It was my first time ever taking a formal class in any kind of programming or computer science – my only regret now is that I wish I had started studying this sooner! Even if I do not end up in a Data Science-related job, these skills are nonetheless enormously useful.
Now that the course is done, I’m back to where I was 3 months ago – unemployed and back on the job hunt. I’m scheduled to meet with a recruiter today, so hopefully something good pans out! Let’s hope. My goal is to hopefully get a job in the intersection of this and chemistry – ideally in cheminformatics, or using Machine Learning models in drug discovery. Even Analytical chemistry positions would not be too bad – these programmatic data analysis methods can be readily applied in that area too. If that does not work out, then I’m considering applying to Master’s programs in CS. This is a really fascinating field, and I would like to get a better foundation in this area.
But yea, now that the course is done, there will be more posts here! Watch this space…