How large would a plain text file be of the entire human genome?

[Gigantopithecus]

Assume the genome is 3,000,000,000 letters long.

 

[MrDudeMan]

whats with all the obvious homework questions recently? do your own work people.

 

[JMWarren]

I dunno, but word doesn't like it when you try to copy and paste it out....
Apparently Word has a page maximum...who would have known!

 

[bobsmith1492]

Well, so you know 1 letter obviously = 1 byte (right...?), and I leave the conclusion to you. Remember 1KB = 1024 bytes and so on and so forth...

 

[Gigantopithecus]

Actually, it wasn't for a homework assignment (stay in school long enough & they stop giving you homework). I have the sequences of the chromosomes downloaded from Project Gutenberg, but the files only total 400mb, which I thought was way too small. Apparently I was right, if 1 letter = 1 byte, then 3,000,000,000 letters should be just under 3gb, right?
(3,000,000,000/1024 = 2929687kb/1024 = 2861mb/1024 = 2.79gb)

 

[JMWarren]

http://www.google.ca/search?hl=en&safe=...n+bytes+in+gigabytes&btnG=Search&meta=

 

[bobsmith1492]

Compression?

 

[Gigantopithecus]

No, they're the draft sequences, not the finished maps (the drafts didn't include non-coding regions).

 

[spidey07]

one byte would equal one pair. Not letters.

Even then you could fit multiple pairs in a single byte. You can do the math. one byte = 256 combinations. divide that by the number of pair combinations and you have how many pairs you can fit in a byte.

 

[eLiu]

I assume they don't put these into a straight text file. For one thing, a standard ASCII character can handle 256 different things. Well, in DNA (let's count RNA too for kicks), there are only 5.

And then sequences of 3 encode single amino acids (or something like that, I don't remember much biology), and there are what... like 20 of those?

So yeah there's tons of room for compression just by shrinking it down this way. Not to mention actual compression algorithms that can take these alphabets and assign them smaller codewords.

 

[dkozloski]

In any case, it's going to be some boring reading.

 

[gsellis]

Originally posted by: spidey07
one byte would equal one pair. Not letters.

Even then you could fit multiple pairs in a single byte. You can do the math. one byte = 256 combinations. divide that by the number of pair combinations and you have how many pairs you can fit in a byte.

2 bits would be all you needed as raw (uncompressed). You have 4 possible answers. A, T, C, G. If A = 00, T = 01, C = 10, and G = 11, 2 bits. 4 pairs per byte. Remember that you do not need to store the pair. You only need one side and the sequence.

 

[Gibsons]

Originally posted by: eLiu
I assume they don't put these into a straight text file. For one thing, a standard ASCII character can handle 256 different things. Well, in DNA (let's count RNA too for kicks), there are only 5.

And then sequences of 3 encode single amino acids (or something like that, I don't remember much biology), and there are what... like 20 of those?

So yeah there's tons of room for compression just by shrinking it down this way. Not to mention actual compression algorithms that can take these alphabets and assign them smaller codewords.

It's available as plain text or a zip file, divided up by chromosome. hope this link works... Text

 

[CycloWizard]

Originally posted by: gsellis
2 bits would be all you needed as raw (uncompressed). You have 4 possible answers. A, T, C, G. If A = 00, T = 01, C = 10, and G = 11, 2 bits. 4 pairs per byte. Remember that you do not need to store the pair. You only need one side and the sequence.

:thumbsup:

Divide the total number of bases by two (only need one side of the sequence), then multiply by two (two bits per base).

 

[phisrow]

ftp://bio-mirror.net/biomirror/genbank/genomes/H_sapiens

 

[Googer]

I do know that if it were put on cd rom and placed in standard plastic cases that the stack would reach over 20+ feet in the air. The Museum of Science and Industy in Tampa has a display of CD cases in the shape of a double helix as a demonstartion of it's size.

 

[JHutch]

Originally posted by: Gigantopithecus
I have the sequences of the chromosomes downloaded from Project Gutenberg, but the files only total 400mb, which I thought was way too small.

Are you sure you downloaded everything? Chromosome 1 is 273.78 MB alone. Chromosome 2 is 245.99 MB, etc ... Quick glance shows that each chromosome averages about 200MB. Plus there is a separate X and Y sequence. Quite a bit more than 400MB of data there...

See http://www.gutenberg.org/etext/3501 thru 3524 for Gutenberg files.

JHutch

EDIT - Granted these are uncompressed text files. Simple zip compression would make it MUCH smaller.

 

[Googer]

Originally posted by: JHutch

Originally posted by: Gigantopithecus
I have the sequences of the chromosomes downloaded from Project Gutenberg, but the files only total 400mb, which I thought was way too small.

Are you sure you downloaded everything? Chromosome 1 is 273.78 MB alone. Chromosome 2 is 245.99 MB, etc ... Quick glance shows that each chromosome averages about 200MB. Plus there is a separate X and Y sequence. Quite a bit more than 400MB of data there...

See http://www.gutenberg.org/etext/3501 thru 3524 for Gutenberg files.

JHutch



EDIT - Granted these are uncompressed text files. Simple zip compression would make it MUCH smaller.

It would, but WinZip does not work on real human genes.