Thursday, March 13, 2014

My Favorite Science Reads - updated

Here is a running list of my favorite science reads (in no particular order), which I will attempt to keep updated. I have included links to the Amazon page as well as the related post from my blog.

The Immortal Life of Henriette Lacks - Rebecca Skloot (my post here)

The Violinist's Thumb - Sam Kean (my post here); The Disappearing Spoon was also excellent (my post)

The Philadelphia Chromosome - Jessica Wapner (my post

Lab Girl - Hope Jahren (my post at CrossTalk)

Inheritance: How Our Genes Change Our Lives and Our Lives Change Our Genes - Sharon Moalem (my post)

The Panda's Thumb - Stephen Jay Gould (all the books I have read by Gould have been excellent)

The Shadows of Forgotten Ancestors -  Carl Sagan and Ann Druyan

The Selfish Gene - Richard Dawkins (The Blind Watchmaker is also great)

Radioactive - Laura Redniss (my post

Blueprint for a Cell - Christian DeDuve 

The World Without Us - Alan Wiesman (my post here)

Stiff - Mary Roach (I review Packing for Mars here, but it wasn't as good)

Tuesday, March 4, 2014

Matt Ridley's Genome - 14 years later

Genome - The Autobiography of a Species in 23 Chapters  by Matt Ridley came highly recommended on Amazon. Despite the fact that the book was published in 2000, I decided to read it. The structure of the book is rather clever: each chapter focuses on one chromosome. This format allows Ridley to expand the metaphor that the human genome is a book: each chromosome is a chapter; the genes are the sentences; the codons (the three letter DNA sequences that the cellular machinery reads) are the words; and the DNA nucleotides are the letters. This kind of metaphor can help non-scientists visualize and remember the organization of the genome. Unfortunately, Ridley eschews the proper scientific terminology, calling codons "words" and nucleotides "letters". Science writing should teach people more about the subject matter; to eliminate the use of fundamental terminology seems a folly.

In each chapter, the author highlights one gene of interest on the chromosome. For example, the chapter on chromosome 13 introduces BRCA2. Like the Jeff Wheelwright book I reviewed previously, Ridley discusses the population genetics of BRCA2. In some chapters, he writes about genetic lessons from a particular chromosome. In some cases, the conceit works brilliantly (e.g., the chapter on the X and Y chromosomes examines sexually antagonistic genes), but other times, it was not as successful. Ridley uses Chapter 21 to discuss eugenics. While the history of the topic is interesting, the connection to chromosome 21 seems a bit tenuous. Parents are increasingly using prenatal screening to detect chromosomal abnormalities, the most common of which is Down syndrome, which is caused by an extra copy of chromosome 21. I found the book worked best when there was an obvious candidate gene to discuss on the chromosome.

It is amazing to consider how much the field of genomics has changed since the publication of Ridley's book (February 2000). In June of that year, the initial rough draft of the human genome was published. When the sequence was officially completed in April 2003, the final cost was estimated to be $2.7 billion; the project took more than a decade. Today, we are approaching the benchmark of the $1000 genome (discussed here and here). The next step is to improve the speed and portability of sequencing equipment as well as the ability to process and analyze the data.

The human genome project has been considered a success in terms of yields for basic research. However, there has been some disappointment that the completion of the project hasn't led to more clinical applications (for more specifics, see these editorials from Francis Collins and Craig Venter, the heads of the Human Genome Project). One problem is that very few diseases are caused by a single gene. Another confounding factor is that there is 1-3% difference between any two individuals' genome sequences. These variations can complicate genomic analyses. To perform a genomic study of a population with a genetic condition, knowing the differences that are normally present in the genome can help narrow down the possible regions that are linked to the condition of interest. Thus, having more complete sequences available will allow scientists to connect DNA sequences with genetic conditions. The 1000 Genomes Project plans to identify all the genetic variations present in the human population. The initial phase of this project was completed in just over four years with the publication of 1,092 complete human genome sequences from 14 populations across the globe. In the coming years, the project plans to complete a total of 2,500 genome sequences to improve the representation of various human populations across the globe. 

Another major change in the genomic landscape is the advent of personal genomics, such as 23andme. These companies offer DNA analysis service that supplies information on your possible ancestry as well as information about other genetic markers. These markers include both the innocuous ones, like whether you can detect a terrible smell in your urine after eating asparagus, and genes linked to possible health risks (e.g., BRCA, Huntington's disease). After intervention by the FDA, the service is now limited solely to ancestry. Scientific American had a fascinating piece about why we should really be concerned about companies like 23andme (TLDR: they are collecting and storing your most personal data
your DNA).

I would not recommend this book to a non-scientist. If you want to learn more about the human genome and DNA, look elsewhere. I highly recommend The Violinist's Thumb by Sam Kean; the recently updated Double Helix by James Watson and the newest book from Craig Venter are likely to be good reads. 


Want
* The Human Genome at 10 Special Issue in Nature covers the changes in the genomics landscape with editorials and articles by a number of major players in the field.
* There is also abundant information on other "big science" approaches to genomics, such as  the HapMap, the ENCODE project, and The Cancer Genome Atlas [TCGA]. I have not explored these for the sake of brevity.