A Crack in Creation; Jennifer A. Doudna, Samuel H. Sternberg; 2018; Mariner Books;246 pgs; Notes, index

​If you examine the book cover, you will notice that some of the letters are in yellow, the others in white. The letters in yellow are A, C, T, and G… the first letters of the four amino acids that make up DNA. The author's last name is "Dou DNA",  but that is just a coincidence. So this is a book about biology. I know a lot more about physics and computing than I do about biology, but oddly, that knowledge was quite helpful. My thanks to Blythe Nilson who corrected some ugly errors in my first draft of this piece.

CRISPR, the underlying technology here, is as important to our future as nuclear power. So you should know something about it. You don't need to know how it works, just what it might be able to do.

Physics and chemistry are closely related. Biology is chemical. And DNA biology is also very computational. DNA's ACGT structure  is binary (i.e.  base 2) code. You might say "Then why are there four letters, rather than two?" Because nature needed an easy way to copy DNA. The DNA chain is made up of paired letters: A always pairs with T, and G with C (each pairing is a bit). This allows the DNA to be cut in two long pieces, and then each piece is reassembled into a complete DNA chain by re-pairing (or repairing) the chain by adding the appropriate matching letter. One copy of DNA becomes two. This is the chemical basis for reproduction.

Each triplet of these letters code for a particular amino acid, and the sequence of DNA dictates how these amino acids will be built into proteins. Proteins are the true stuff of life.

As I read this book, I was struck by the number of times I could see software analogies in the chemistry. In computing, data and code are two sides of the same coin. The same is true for the molecules of life. They are hardware (a fixed bunch of atoms arranged just so) and software (do this, then do that) all packaged up into a single object.

Enter CRISPR (an acronym for Clustered, Regularly Interspaced, Short Palindromic Repeats), a new technology that is both enormously promising and bloody scary at the same time.  CRISPR is not a great name. Even knowing what the letters stand for tells you nothing about what is actually is. The one chapter in the book that describes the CRISPR details is challenging. It is full of acronyms and strange words, making it hard to follow. The rest of the book is much less challenging.

CRISPR was initially just an observation that part of the DNA of the bacteria consists of Repeating Clusters of DNA. The repeated bits read the same forward as backward (Palindromic), were quite Short, and were always the same distance apart (Regularly Interspaced). It soon became apparent that these genes were associated with the bacteria's immune system. What a bacteria fears is a phage (short for bacteriophage), a virus that attacks bacteria. The CRISPR genes contain a length of genetic material in the Regularly Interspaced part. These bits of RNA are actually viral RNA that was taken from a phage  in the past, and is now used as a pattern matching template to recognize viral DNA. Associated with CRISPR is an enzyme that, once activated, destroys its DNA/RNA target… in this case, the phage. 

Aside: RNA and DNA are chemically very similar and sometimes serve similar purposes. RNA is a single stranded molecule, and DNA is double stranded. RNA uses uracil (U) instead of thymine (T) in its code.

Once the enzyme is released, it zooms down the DNA chain at a rate of 300,000 nucleotides per second, carving it up into amino acid junk! That is fast!

Scientists realized that this mechanism could be used to find, change, and/or disable genes with amazing accuracy. For a software guy like me, I see many analogies to computer code. Each segment of phage DNA/RNA in CRISPR is used as template to find an invading phage and kill it. This is like a parameter to a subroutine or, if you prefer, a kind of microscopic Google search using the DNA segment as the search target.  This is hardly surprising since, at its core, genes are a series of zeros and ones that are used to make you and me. In other words, it is all software and software is easy to change (hence the "soft" part).

CRISPR technology has enormous potential for both good and evil. It might be used to cure horrible genetic diseases such Huntingtons, or it could be used to create supermen. It can be used to hunt down  one gene with one wrong letter, tag that gene for repair, and then get the mechanisms of body to repair it. Sickle Cell Anemia is one such disease.

When writing software, one generally designs top-down and builds bottom-up. In the software of real life, there is no design,  only what works. Software starts with building tools; and then uses those tools to build larger software structures which, in turn are used as tools to build even more complex structures and procedures. And with all these tools lying around, there often comes a realization that the existing tools could be easily repurposed to do something  that previously seemed out of reach. I have experienced this many times in my software career. Bio-researchers are discovering all these tools lying around in the cell and are closing in on learning how to use them.

CRISPR opens many doors, some of which we should probably keep locked. Curing an awful disease is obviously a good thing. Changing human germ cells is much scarier. Changes to human germ cells means that the change is passed on to offspring. And that smacks of eugenics, NAZI supermen, designer babies etc., and it raises many ethical questions.

The closing chapters of the book focus on the future and the inherent advantages and dangers that CRISPR embodies.        
 
Biotech like CRISPR gives us god-like powers to manipulate life. Advances in biotech and computing make it possible for us to wield those powers.  I do not think it an exaggeration to say that CRISPR is the biotech equivalent of the Manhattan Project. I hope mankind learns to use it wisely, because use it we will. 

---