Sequencing is the new microscope

Preamble

One of my biggest personal fears is working in the wrong field to achieve the goal I care about. If you were around pre-1900s, and wanted to contribute to biology, you should have been a physicist (Robert Hooke, a physicist discovers the first cell, making a better microscope is a major driver of progress). In which field should you work to maximize progress in biology today?

Tools are fundamental to progress in science. Many of the best scientists in history made their own tools. For example, Isaac Newton was one of the foremost experimentalists of his time - the most vivid theme permeating his notebooks is the constant creation of new tools. Other examples include Louis Pasteur, Edward Bluchner, and Michael Faraday with his own legendary notebooks. These scientists didn’t just ask questions and extrapolate based off of existing data - they were excellent toolmakers. They create new experimental apparatuses to answer the questions that came to mind.  

This is biology’s century

For the longest time, looking at stuff and breeding animals was effectively our version of biology. That and some pretty uneducated guesses about what human organs did. 

That changed in the 1600s - microscopes allowed us to see all the way down to the cell. The first revolution in physics gave us optics (we could see a lot more stuff a lot better), and Newton’s laws of motion (we could centrifuge and pipette things accurately). Understanding physical principles in the second physics revolution - electromagnetism and thermodynamics - allowed us to run western blots (from the principle of charge) and make incubators that could cool as well as heat

As the progress of physics ramped up in the early 1900s, so did biology. JJ Thomson’s device for looking at cathode rays became the first mass spectrometer. X-rays were found - now used, not just in hospitals for treatment and diagnosis, but also as invaluable biological reagents. NMR gave rise to MRI. Einstein’s general relativity didn’t make much of a dent, but the photoelectric effect allowed us to understand and manipulate fluorescence. Microscopes got way better - the Nobel in Physics was won for phase-contrast microscopy, then electron microscopy. Marvin Minsky patented the first confocal microscope.

The flywheel of biological tools 

But something interesting happened around the 1950s. If you look at the most important techniques in biology, in the second half of the 1900s, they’re all driven by tools discovered in biology itself. Biologists aren’t just finding new things - they’re making their new tools from biological reagents. PCR (everything that drives PCR, apart from the heater/cooler which is 1600s thermodynamics, is either itself DNA or something made by DNA), DNA sequencing (sequencing by synthesis - we use cameras/electrical detection/CMOS chips as the output, but the hijacking the way the cell makes DNA proteins remains at the heart of the technique), cloning (we cut up DNA with proteins made from DNA, stick the DNA into bacteria so living organisms can make more copies of it for us), gene editing (CRISPR is obviously made from DNA and with RNA attached), ELISA (need the ability to detect fluorescence - optics - and process the signal, but antibodies lie at the heart of this principle), affinity chromatography (liquid chromatography arguably uses physical principles like steric hindrance, or charge, but those can be traced back to the 1800s - antibodies and cloning have revolutionized this technique), FACS uses the same charge principles that western blots do, but with the addition of antibodies. 

There are some exceptions. Chemical peptide synthesis might be one. Microscopy is still in large part driven by physics on a certain level - Cryo-EM and lattice light sheet microscopy were iterations on known physical principles (albeit longstanding known ones). But even there the analogy falters. If you’d asked a physicist in the early 1900s how we’d figure out the complicated, tangled structure of DNA, he’d probably say to look at it with a microscope. Sequencing has become the new microscope. It’s easier for us to cross-link, fragment and sequence a full genome to figure out its 3-dimensional structure than it is for us to figure that out by looking at it head-on. We used to rely on photons and electrons bouncing off of biological sample to tell us what was going on down there. Now we’re asking biology directly - and often the information we get back comes through a natural biological reagent like DNA. Which we then sequence using motors made from the DNA itself! 

These tools don’t just find new biology, they’re also being used to change it. Look to the market, where biology-driven therapeutics are coming out in droves. The first generation of drugs came from natural products - for the most part things created by living organisms, but more approachable from a chemistry perspective. A 100-atom structure could come from a living thing, but also be created through a set of chemical reactions. But then, in the 1980s, we cloned insulin and growth hormone, and used living things to make the drugs that previously had been isolated by purification. Antibodies became the wonder-drugs of tomorrow. These are complex 100K-atom molecular structures, so we didn’t make them - we let cells and bacteria do that work for us. Viruses similarly - we used them as tools to understand genetics, now we're using them to blow up tumors (look up 'oncolytic viruses') and change our genes. We're using biological tools to program and deploy cells to fight cancer and autoimmune disease. In the future, one can imagine taking advantage of an even more complex level of biological hierarchy - interfacing with and using our nervous system (and here I turn it over to the enthusiasm of the BCI folks). 

But the point isn’t just the obvious - that this is biology’s century, that we’re seeing an incredible wave of drugs come to market. The point is that something truly unique is happening in biology today. 

Sequencing is the new microscope

Something special happens when a field becomes self-reinforcing. Previously, biology looked to physics and other disciplines for tools to break open new frontiers. But, empirically, since the 1950s, that has all changed. We don’t make mutant mice with x-rays and microscopes - we figure out the gene we want to go after, and we use high-precision biological tools to change it. Computer science has certainly played an important role in processing all of the information now streaming out of biological systems, but the major advances - the core things driving progress in biology forward - have come from biology itself. Biology is eating physics (and, some would jokingly suggest, based on the outperforming endurance of DNA compared to any modern hardware and the plausibility of biological computation, possibly computation itself). 

Naively, if we can expect n new discoveries / t tools we have, if the tools are static, maybe that’s a fixed number of discoveries per year. But if the number of tools increases, then we get more discoveries. What if that number increases as a function of n?

This is important because it’s a self-reinforcing loop. The more things in biology we discover today, the faster we can discover things tomorrow. Biologists are the new engineers. But their tools look a lot different than any we’ve seen before. Sequencing is the microscope of tomorrow. And sequencing was built by biological tools. 

views
33 responses
I worked on the human genome developing shotgun sequencing. Today we need to train students in DNA sequencing so they can create the next set of DNA tools. NCGR trains college students http://www.nminbre.org/index.php/2018-inbre-sum... I created a Mu Python program to simulate MinION. https://www.youtube.com/watch?v=9tUtC4rcZ3w Mu Python plotter can use ACTG tuple to simulate MinION. MinION is now used for T-Cell and B-Cell characterization by reading immune response to breast cancer. https://www.biorxiv.org/content/biorxiv/early/2... Nick Lowman has some great intro to MinION http://lab.loman.net/​ http://lab.loman.net/2017/03/09/ultrareads-for-...
This python code was created with python Mu editor for students to begin building python3 skills. https://codewith.mu/en/tutorials/1.0/python DNA sequencing uses 4 base pairs (ACTG) to create a genome. This python model will use Mu to record base pairs in a simulated DNA sequence from a next generation DNA sequencer. The following model of an Oxford nanopore DNA sequencer will show what a Mu DNA plotter simulation may look like. https://www.youtube.com/watch?v=GUb1TZvMWsw My son used accelerometer with GPS to plot our car to college. He captured the data while we drove a car. This is a good Mu plot model. https://www.youtube.com/watch?v=WTCCzxT6kqQ https://codewith.mu/en/tutorials/1.0/adafruit https://codewith.mu/en/tutorials/1.0/plotter Python is the most common language taught in college and used in software jobs in US. This DNA plotter can be used as a beginning program for students. Install Mu https://codewith.mu/en/download By using the Mu plotter students can watch the transitions from A to T to C to G bases. A tuple is used to provide the Mu plotter with a set of data to plot. This is similar to a MinION DNA sequencer recording DNA A tuple is a sequence of immutable Python objects.Tuples are sequences, just like lists. The differences between tuples and lists are, the tuples cannot be changed unlike lists and tuples use parentheses, whereas lists use square brackets. Creating a tuple is as simple as putting different comma-separated values. Python3 DNA Base plot code import time import random base=1; while base: # Configure outputs from bases time.sleep(0.05) #Adenine(A) if base==1: gp1=(53,0,0,0) #Tuple of Adenine (A) print(gp1) time.sleep(0.05) #Thymine(T) elif base==2: sp1=(0,50,0,0) #Tuple of Thymine (T) print(sp1) time.sleep(0.05) #Cytosine(C) elif base==3: wp1=(0,0,45,0) #Tuple of Cytosine (C) print(wp1) time.sleep(0.05) # Guanine(G) elif base==4: gt1=(0,0,0,35) #Tuple of Guanine (G) print(gt1) # Ask user which base to use base=int(input("1 A 2 T 3 C 4 G?")) print ("Your Base is:",base) type(base) Have students build a DNA sequencer model with user interacting with MinION data. The use of tuples to populate a CSV file for use in a database or in machine learning. https://github.com/kjaisingh/high-school-guide-...
31 visitors upvoted this post.