Understand Your DNA

A Guide

Lasse Folkersen

• Publisher: WSPC
• Year: 2018
• ISBN:
  9789813273276
• Topic: Scienze biologiche
• Subtopic: Genetica e genomica

Are you considering to test your own DNA? Do you want to learn more about your health and ancestry? Understand your DNA — A Guide is about what you can use genetics for. For a few hundred dollars, you can now scan your own genes. Millions of people all over the world have already done so. Everyone wants to see what they can get to know about themselves, and the market growing rapidly. But what does it require from you? And what can you really use a DNA test for? Understand your DNA — A Guide helps you put the plots and charts of consumer genetics into perspective and enables you to figure out what's up and down in the media headlines. The book is also a key input for today's debate about what we as a society can and want to do with medical genetics. Genetics will play a growing role in the future. Understand your DNA — A Guide is an easy-to-read and necessary guide to that future. The book is provided with a foreword by Professor Sham Pak-Chung of Hong Kong University.

While there are many books about genetics, they typically take the perspective of a scientist wanting to understand the molecular levels. At the same time, direct-to-consumer genetics is a booming market, with millions of people already tested. Very little has been published that will guide them for real, because the need here is more focused on medical and practical understanding, than focussed on molecules.

This book therefore aims to hit that vacant spot in the market. It's a walk-through of all concepts that are necessary to understand in your own analysis. Meanwhile, it is also limited in scope to only those concepts — thus distinguishing it from broader works.

The book is appropriate for the readerships in modern multi-ethnic metropolises because it mixes European and Asian examples, both from the collaboration between the author from Europe and the foreword-writer, Prof. Pak Sham of Hong Kong University. But also, because many of the examples in the book concerns differences and similarities between Asian and European ethnicities, something the author believes is a trend in time.

Contents:

• Foreword
• My Daughter's Hair Colour
• Basic Introduction to Genetics
• Methods in Consumer Genetics
• Ancestry and Genealogy
• Rare Disease Genetics — One SNP at a Time
• Genetics of Common Traits and Diseases — Many SNPs Together
• The Future
• Acknowledgements and Disclosures
• Bibliography
• Appendix A: Operating Instructions for Impute.me
• Index

Readership: Students, General Public, and Others (people who have recently taken direct-to-consumer gene tests).
Key Features:

• Genetics is a highly trending subject in technology, particularly right now with the booming of consumer genetics
• The treatment of genetics from the point of view of consumer genetics is unique
• There is a prominent foreword writer, Prof. Pak Sham of Hong Kong University

My Daughter’s Hair Colour

Some children look like their mother, some like their father, some are hard to tell. My daughter definitely looks like me. Our hair colours are almost identical, and face shape is similar as well. At least she got her eyes from her mother. I can’t stop marvelling at this. I’m sure I’m not the only father (or mother) thinking about these matters, maybe even thinking a little too much. But what was a little remarkable is that while my hair colour is an average Scandinavian light brown, her mother’s hair is complete Chinese jet-black in appearance. And this is very unlike the genetics you learn in high school. Here you may have heard that dark hair is a dominant trait. That dark plus light gives dark, always. So clearly an exception. Not too uncommon though; this happens all the time. It could have been all there was to it. But because of my deep interest in genetics the exception instead turned me into a long path of analysis, ultimately resulting in this text.

It was puzzling. Sometimes annoying to the point where people would ask my wife “where the mother is”, that is until the two of them started chatting in fluent mandarin. A puzzle, but potentially a solvable puzzle. So, I set out to get DNA samples from my daughter and all her living ancestors, ten people altogether, including myself and my wife (Fig. 1.1). I had decided to use 23andme because back then in 2012, it really was the most well-known provider. Plus, they offered a toddler sampling kit that could be used for sampling. Spit collection from a two-year old can be remarkably difficult, in spite of what you may think from the drooling. Collection of spit from elderly Chinese in-laws can give some explanation problems as well, but they are used to me doing odd things.

Fig. 1.1.Proband and pedigree of participants in the example study. Proband is a word that basically means “the one the genetics study starts from”, so in this case my daughter and her hair colour. For privacy reason, I have not given actual names, but instead used local-language family labels. Their meaning should be clear from the position in the pedigree, but they are anyway not important to know, except “Far” who is me and my daughter who is indicated. Per pedigree convention, female is indicated as circle and male as square. A crossed-out symbol refers to a deceased individual.

Genetic testing of your entire family is of course something that needs to be well planned and well explained. First, I made an agreement with everyone in the family that I would use the genetic data only for non-medical investigations, with the exception that later medical analysis was possible but would require a thorough talk and mutual agreement.

Secondly, genetic testing is inevitably also a test of parenthood; if the father is not who you think it is — you find out from genetics. I had decided that if I suspected this, it would be a major no-go. For the sake of the family, better to just cancel the project than shake up marriages. From large-scale studies that I have been involved in we know that this is very real and that 3% to 5% of fathers are not the actual biological parent. Of course, the real challenge is to figure out if this may be a problem before collecting the samples. The way I did it was to take a serious private talk with all the mothers involved (my own mother included), clearly saying that one of the things we would find out was the true paternity. Then pausing a little and changing the subject to something more medical, about if they were afraid of finding out about diseases. I thought that way I could give people an easy way out. And no one took it — indeed all fathers turned out to be biological fathers as well. But it is worth thinking about if you consider doing a similar investigation.

After sampling the spit and sending it off, it usually takes several weeks before the data arrives. Once that happens, the first thing you get is a page on a website with information about specific traits. This is true no matter what company is used. Naturally, I zoomed in on the hair colour trait as the first thing, and found little new information — my side of the family was predicted to be of light hair colour and my wife’s side was predicted to be of dark hair colour. My daughter was predicted as somewhere in between. However, on digging a bit this opened up the whole problem with many analytics interfaces: it’s based on old research, incomplete findings and simplistic assumptions. In this case hair colour was calculated as a function of just two SNPs. A SNP, pronounced “snip”, is the functional genetic variant and we will cover what it is in much more detail later. However, only basing something as complex as hair colour on two SNPs was clearly not true per the latest research on multi-genic traits. Overall, a good rule of thumb in the world of genetics self-analysis is this: be suspicious of the sources. Newer research is usually more comprehensive, and common traits are almost never decided by just one SNP.

Fortunately, deeper digging can amend this. Almost all companies, 23andme included, allows the download of your raw data. Your raw data file is a large text or excel file that contains one line for each SNP, typically around a million lines. Not something that can readily be processed manually, but very useful to get deeper information from. In the case of hair colour, it turned out that a later study (Eriksson et al., 2010) had found more than 22 SNPs that determined hair colour, on both the blond-to-black scale and on the red-to-not-red scale. These SNPs would be a place for a correct answer to be found. So I started to find the genotype information for each of these 22 SNPs in the data from each of my ten participants, including my daughter.

The most decisive effect was a SNP called “rs16891982”, which is known to have very strong effect on hair colour and which completely corresponded to ethnicity. All the Chinese family members had C/C and all the European family members had G/G. This is generally true for that SNP, not just in my family. The C is only found in people of Asian ethnicity. My daughter, of course, had C/G. This is, of course, because of the way inheritance works — everyone receives one copy from their mom and one copy from their dad. I could only give G and my wife could only give C, so C/G was the only possible result for our daughter. We will discuss these inheritance patterns in more detail in Chapter 4. C/G is also the rs16891982 genotype of all other children with one Asian parent and one European parent, so clearly it is not the explanation for an unusually light hair colour in my daughter.

But what about the 21 other hair colour associated SNPs that were known? These could contain clues. It turned out that the European side of the family, including myself, had a fairly average mix of genotypes in these SNPs. Some were of the lighter variant, some of the darker variants. This makes sense. We are not platinum blondes, just fairly average brown Scandinavian hair colour. But when I extracted these genotypes from my wife and the Asian side of the family I found that remarkably many of them were of the lighter variant. My wife’s family seemed to be genetic blondes, so to speak. Looking at a family photo, this is surprising. They all have jet-black hair. It illustrates the concept that we will discuss later, the difference between phenotype — how you look — and genotype — how your DNA looks. So, even with the 21 other hair colour determining SNPs being fairly light variants, it makes sense that the rs16891982-Asian-hair-colour-SNP trumps them and creates a completely black hair phenotype. This is why most Chinese have black hair.

This trumping effect — or dominance — however was not present in my rs16891982-C/G daughter. And having received an otherwise very blonde genetic “palette” from her mom, as well as a medium blonde palette from me — this was the explanation for our mystery. My wife — a genetic blonde — would, together with me, have a much higher chance of having light-haired children than what is average for European-Asian couples.

As a good scientist, I am currently arguing for us to repeat the experiment a few times, but that is a family matter that is not agreed yet — so far we have only one child — and so, until further replication I consider this the answer to the originally posed question. My wife is a genetic blonde.

This also sets the aim of this book: to cover everything that you can do with a personal genome. Your personal genome is the information on the A, G, C and T’s that make up your DNA code — your book of life. A short basic introduction on this is given in Chapter 2, a basic introduction to molecular biology. The data for a personal genome typically comes from genetic tests sold by such companies as 23andme, Ancestry.com, MyHeritage, or Family Tree DNA, all collectively called direct-to-consumer (DTC) genetics testing companies. But it may also have come from research projects, e.g. Genes for Good, or through clinical testing in a hospital. Chapter 3 will give more detail on the current approaches and methodologies for how this is done. Most consumer companies provide specific analytics interfaces, most with specific scopes and aims, for example ancestry. Adding additional analysis will enable you to learn more. This is the reason for this book; to provide a handbook in all types of personal genome analytics. Specific analysis types can be grouped into ancestry studies, rare disease genetics, and common disease genetics. Each of these themes will be introduced in Chapters 4, 5 and 6 respectively, conveniently sorted by how much mathematics it requires to understand each.

Ultimately, the hope with writing this is in part an encouraging pitch towards educating yourself in personal genetics. We can learn many things from our genes, deep questions about our ancestry, curiosities about hair colours, but also medically useful insights. But it is also an emphasis on limitations of interpretations; our genes are not a deterministic crystal ball that have the complete answer to everything. To gain any form of benefit or insight from genetics, most of all it is important to understand this gap between no interpretation and over-interpretation.

Basic Introduction to Genetics

The human genome consists of more than three billion DNA letters: A, C, T or G. They are, of course, not real letters, but four different kinds of molecules, nucleotides, that we have named as A, C, T and G. Strung together, they encode all of our genes, and can very well be thought of as letters in a book, three billion of them in length. Because it is these genes that make all the proteins that exist and make the things that we consist of. So, it is very true when people describe the DNA as the blueprint for an organism, or book of life. Everything that we consist of can ultimately be traced back to information given in our DNA.

Many people have heard about the sequencing of the human genome around the turn of the century, and the immense efforts involved in figuring out the more than three billion letters of our DNA. After the efforts of the first human genome project, scientists started to ask what the DNA of all the other people in the world looked like. Thousands of genomes from people all over the world were DNA-sequenced, in a project named the Thousand Genomes Project. And it turned out that, by and large, humans were fairly similar genetically. In fact, at 99.9% of the DNA positions, almost all humans had the exact same letter, the same nucleotide.

2.1.SNPs, Genomes, and Chromosomes

To understand better what this means exactly, let us zoom in on a specific place in the genome. We will zoom in on a small region containing a SNP called “rs16891982”, the one we discussed in Chapter 1. The concepts are general for any region of the genome, though.

Fig. 2.1. Sequence of 45 nucleotides on chromosome 5 in five organisms, three of whom are human. This region is part of the SLC45A2 gene. The center part highlights the rs16891982-SNP.

In this particular region, the nucleotide code is as shown in Fig. 2.1. Long stretches of A, T, C and G. It continues like that in each direction, millions of letters. The central highlighted letter, the one that is not identical in all of the genome sequences, that is a SNP — one of the positions of known variability in humans. SNP is an acronym for Single Nucleotide Polymorphisms and it is pronounced snip. The only reason we call this position a SNP is because it is not always identical; different humans have different letters at this position. Because of projects like the Thousand Genomes Project, all SNPs in the human genome are known.

For a position to be called a SNP, that position must have been observed to have different letters in a least some individuals. SNPs are recognisable and distinguishable because they all have names beginning with rs plus a number, for example rs16891982. We have named them like that because we wanted to categorise them and keep track of them — which SNPs are associated with which traits.

Together with the SNP name, a chromosome number and a position is often given. The region in Fig. 2.1 is located at a position given as letter number 33,951,943 on chromosome 5. Chromosomes are a level of organisation of the genome, where all the DNA sequence is physically divided into 22 ordinary chromosomes, as well as X and Y chromosomes that determine the sex. They can be loosely thought of as volumes in a book, but other than the sex chromosomes, there is nothing much special about a location on a specific chromosome. Most traits are scattered around chromosomes anyway.

A SNP may be located in a gene or next to one. A gene is a functional region of the DNA that is read and translated into the constituents of the body: proteins and the products that they in turn make. How that happens is the subject of the entire field of molecular biology. Like the term chromosomes, however, genes actually play only a secondary role when analysing your own genome. Because the functional unit that is measured is the SNP. SNPs are important. But now you also know the terms gene and chromosome, and they are nice to know.

2.2.The Alleles that Make the Genotype of a SNP

The next thing you’ll notice in Fig. 2.1 is that each individual has two sequences of DNA. This is a consequence of sex. We all inherit one copy of our genome from our mother and one copy from our father. Of course, your mother and your father also had two copies each, so there’s a bit of random shuffling in which version they’ll pass on. Basically, this is the choice of which egg cell and which sperm cell is involved in conception. Quite an important moment in anyone’s life. The specific combination of letters that you have for a given SNP is what we call your genotype.

The word allele is the word we use to describe what letters the genotype consists of. It just means a letter, in the context of another possible letter. So, you can say for example “the C-allele” to distinguish from having a G-allele at that genomic position. But it will sound odd if you say “C-allele” at a position that always have C in all humans. I ...

Table of contents

• Cover
• Halftitle
• Title
• Copyright
• Dedication
• Contents
• Foreword
• 1. My Daughter’s Hair Colour
• 2. Basic Introduction to Genetics
• 3. Methods in Consumer Genetics
• 4. Ancestry and Genealogy
• 5. Rare Disease Genetics — One SNP at a Time
• 6. Genetics of Common Traits and Diseases — Many SNPs Together
• 7. The Future
• Acknowledgements and Disclosures
• Bibliography
• Appendix A Operating Instructions for Impute.me
• Index