Blog Article

How Much Information is Stored in the Human Genome?

by on 16th of March, 2012 in Inspiring & Thought Provoking
Grigoryev_header image copy

The other day I was having a conversation with a friend of mine who had some background in computer science. The conversation shifted towards my research and the following question came up: What is the amount of digital information stored in a human genome? I started searching in the deep dark corners of my brain, but I realized that I simply did not know the answer. So I decided to do the math to estimate how much information is stored in our genome.

Laying out the information storage capacity of the genome

The human genome contains the complete genetic information of the organism as DNA sequences stored in 23 chromosomes (22 autosomal chromosomes and one X or Y sex chromosome), structures that are organized from DNA and protein. A DNA molecule consists of two strands that form the iconic double-helix “twisted ladder”, whose backbone, which made of sugar and phosphate molecules, is connected by rungs of nitrogen-containing bases. DNA is composed of 4 different bases: Adenine (A), Thymine (T), Cytosine (C), and Guanine (G).  These bases are always paired in such a way that Adenine connects to Thymine, and Cytosine connects to Guanine.  These pairings produce 4 different base pair possibilities: A-T, T-A, G-C, and C-G. The haploid human genome (containing only 1 copy of each chromosome) consists of roughly 3 billion of these base pairs grouped into 23 chromosomes. A human being inherits two sets of genomes (one from each parent), and thus two sets of chromosomes, for a total of 46 chromosomes, representing the diploid genome, which contains about 6×10^9 base pairs.

Comparing the genome to computer data storage

In order to represent a DNA sequence on a computer, we need to be able to represent all 4 base pair possibilities in a binary format (0 and 1). These 0 and 1 bits are usually grouped together to form a larger unit, with the smallest being a “byte” that represents 8 bits. We can denote each base pair using a minimum of 2 bits, which yields 4 different bit combinations (00, 01, 10, and 11).  Each 2-bit combination would represent one DNA base pair.  A single byte (or 8 bits) can represent 4 DNA base pairs.  In order to represent the entire diploid human genome in terms of bytes, we can perform the following calculations:

6×10^9 base pairs/diploid genome x 1 byte/4 base pairs = 1.5×10^9 bytes or 1.5 Gigabytes, about 2 CDs worth of space! Or small enough to fit 3 separate genomes on a standard DVD!

Data storage across the whole organism

Some interesting question could follow. For example, how many megabytes of genetic data are stored in the human body? For simplicity’s sake, let’s ignore the microbiome (all non-human cells that live in our body), and focus only on the cells that make up our body. Estimates for the number of cells in the human body range between 10 trillion and 100 trillion. Let us take 100 trillion cells as the generally accepted estimate. So, given that each diploid cell contains 1.5 GB of data (this is very approximate, as I am only accounting for the diploid cells and ignoring the haploid sperm and egg cells in our body), the approximate amount of data stored in the human body is:

1.5 Gbytes x 100 trillion cells = 150 trillion Gbytes or 150×10^12 x 10^9 bytes = 150 Zettabytes (10^21)!!!

Sexual information exchange

Along the same lines, how much genetic data is exchanged during human reproduction?Each sperm cell in a human male is heterogametic and haploid, meaning that it contains only one of two sex chromosomes (X or Y) and only one set of the 22 autosomal chromosomes. Thus, each sperm contains about 3 billion bases of genetic information, representing 750 Mbytes of digital information. The average human ejaculate contains around 180 million sperm cells. So, that’s 180 x 10^6 haploid cells x 750 Mbytes/haploid cell = 135 x10^9 Mbytes=135000 Terabytes!!!! Following this idea even further, while 13500 Tbytes are transferred, only one sperm cell will fuse with an egg, using only 750 Mbytes of data, combining it with another 750 Mbytes of data from the egg. Thus, essentially 99.9999…% of the data transferred during sexual reproduction is lost in the pipeline … Whether the remaining fraction of information will result in anything constructive is up to good parenting.

Having worked out the above numbers, a whole bunch of other curious questions can be asked. Have you ever wondered about the data capacity of our biological organism? What is the rate of data transmission during cell division? The rate of data transmission during gamete fusion? The rate of data transmission when human lymphocytes circulate through the bloodstream? What amount of data is destroyed daily by apoptosis? What amount of data is created daily?  How does this compare to the rate of data transfer via an optical fiber?

Please feel free to contribute your own dubious calculations and questions below!

Photo Credits
About the author: Yevgeniy Grigoryev
Yevgeniy grew up in New York, but decided to transplant himself to the West Coast for his PhD studies at the Scripps Research Institute, where he studied mechanisms of gene regulation in the immune system. Recently, Yevgeniy again found himself in New York City, pursuing a post-doctoral research project in oncology. Realizing that his true passion was writing, Yevgeniy decided to leave academia and pursue science writing as a freelancer to share his passion for fostering communication between the scientific community and the public. In his spare time Yevgeniy works as a Krav Maga self-defense instructor.

See more from Yevgeniy Grigoryev Twitter

9 thoughts on “How Much Information is Stored in the Human Genome?”

  1. Avatar of Balaclava Balaclava says:

    You are completely ignoring the fact that information in DNA is not only present in the base sequences but also (and maybe more so!) in its methylation profile. So when you are trying to represent a single specific genome in the form of bytes you have to take that into consideration.

    Furthermore, when you are calculating the amount of data stored in the entire human body you have to take into account that not all promoters are in the same state in every cell (that's why every cell is different..!)

    So how would you represent acetylations and methylations?

    1. Dear Balaclava,

      Your observation about the methylation and acetylation profile are very valid, such epigenetic factors make the coding capacity of our genome almost infinite! However, what I tried to calculate was mere data "stored", not expressed by the human genome. My calculations are oversimplified, of course. I doubt that currently there is a way to calculate the amount of data "expressed" in the genome, that also factors in all the genome-wide epigenetic modification events such as methylation and acetylation.

      You are also absolutely right about the varying promoter states across different cells. However, the promoter expression does not affect the data stored. I think it would be virtually impossible to calculate the data actually expressed at any moment, taking into account any all the possible promoter states and epigenetic events. Of course, all attempts at such daring tasks are more than welcomed here, this is the point of this post, after all.

  2. Avatar of Emily Crow Emily Crow says:

    What do you think epigenetics would equate to in computer terms – processing power? Could different promoters be thought of as the amount of operations that can be carried out simultaneously??

  3. So my knowledge of how computer works is rather limited but I think processing power is a good comparison. While the genetic data stored can be compared to RAM (Random access memory, a form of computer data storage). The amount of data expressed can be compared to processing power or computing power. The epigenetics machinery and promoters can be compared to computer processors that handle massive calculations. Alternatively, they can probably also be compared to the CPU's electronic clock, that creates a series of electrical pulses at regular intervals. This allows the computer to synchronize all its components and determine the speed at which the computer can pull data from its memory and perform calculations. In a cell that would equate to all the active transcriptional states at any given time.

  4. Avatar of 1857 1857 says:

    hi, this is leili, i would tahnk you if you add more last seminars and congress in biology, please,,,,,,,,,,,,

  5. Avatar of DougB says:

    I have say much credit is due for even attempting the calculation. How about the data storage capacity of cellular mitochondria? It seems this also would add to the total. I am curious to know how your friend responded to the storage capacity of DNA.

  6. Avatar of chrismckay chrismckay says:

    Your calculations are based on the chromosome, I pose the question would 150 Zettabytes be enough to store the human body as a binary form and then be able to reintegrate it?

    1. Avatar of smartmoves smartmoves says:

      Give or take an order of magnitude, a 1Ghz async network would take 1,750 years to transfer that much data. I guess "beam me up Scotty" is out to lunch for a while…

  7. Avatar of Mike Levens says:

    All of this is a good discussion of the amount of data in the human genome, but it's not really the same as discussing the amount of *information.* Storing an arbitrary 150-zettabyte value as a human would require the possibility of having human beings who have a completely different genome in each and every cell, which is obviously crazy. So how much redundancy is there, and what is the Shannon entropy of the data being stored?

    Suppose my sci-fi future self wanted to write a compression algorithm for storing and transmitting a particular human genotype, say for molecular reassembly and cloning at a remote location or something. What data compression ratio could I achieve in theory? How big would this 150 zettabytes look like if LZW'd?

Speak Your Mind

Article Categories

Inspiring & Thought Provoking