Your question is too vague and too simple to answer in a few sentences.

"Coding" and "non-coding" are technical terms that refer specifically to the translation of messenger RNAs (mRNAs) to proteins; DNA doesn't participate in that process. The DNA bases from which the mRNA is transcribed are referred to as "coding sequence" because through transcription into RNA, they specify (or "encode") amino acid sequences. The process of translation requires RNA molecules that are also transcribed from the DNA, including transfer RNAs which hold an amino acid at one end and contact three consecutive RNA bases at the other end; these transfer RNA "genes" are part of the "non-coding" DNA, and they are needed in such quantity that there are many copies of each transfer RNA in the DNA. There are also ribosomal RNAs (rRNAs), which make up part of the ribosome structure which translates mRNAs into proteins; these genes are also non-coding genes and are present in many copies. There are several other categories of small RNAs that are transcribed from the DNA but are not translated into proteins, including microRNAs (miRNAs) which appear to help determine how often mRNAs are translated into proteins; small nuclear RNAs (snRNAs) which are required for splicing large segments (introns) out of transcribed RNAs, including mRNAs; small nucleolar RNAs (snoRNAs) which, among other things, guide the process of chemical modification of some RNAs, especially the rRNAs. Some of these, especially snoRNAs and miRNAs, are transcribed as part of introns (mentioned above) and then cut out of the intron to become functional. Even though introns seem to be useless, because they are rapidly cut out of an mRNA that goes on to be translated into a protein, it's worth keeping in mind that there may be snoRNAs or miRNAs in that intron that are also beneficial to the cell. We tend to think "linearly", i.e. one part of the DNA has one "purpose", but biology doesn't work that way. The presence of an intron is known to be important to getting a protein produced, in the sense that if you just add a gene to the DNA of human cells in culture, and that gene doesn't contain any introns, it is most likely to be transcribed and translated poorly; but inserting an intron (any intron, it doesn't appear to have to have some special component hidden in it) into that same gene often causes it to be transcribed and translated much more efficiently. So, is an intron sequence "junk" DNA, or not?

However, DNA has many functions other than encoding proteins, rRNAs, tRNAs, snRNAs, snoRNAs, and long(er) non-coding RNAs (lincRNAs)! A very important function of DNA is to be copied and delivered to the next generation, and a lot of the DNA sequence participates in that process. The centromere contains specific signals that are important in separating two copies of a DNA molecule into different cells as a cell divides. The telomeres (the ends of a linear DNA molecule, or chromosome) contain repeated signals that are bound by specific proteins to prevent the DNA from being shortened. DNA is packaged into a structure called chromatin, which requires the DNA to bend around histone proteins; this bending has been shown to be related to the DNA sequence, in the sense that there are weak, but detectable periodic patterns that result from, or contribute to, the bending process. All, or at least the vast majority, of the information that determines which genes are transcribed, and when, and in which cells, must ultimately be recorded in the DNA sequence itself; and most of that information isn't considered "coding". We have a long way to go to figure out which parts of the DNA are involved in which processes.

There are also some very puzzling regions of the genome that aren't transcribed, but they are very strongly conserved over millions of years of evolution. The fact that they are so well conserved suggests they are important for SOMETHING, but it really isn't clear yet what they do.

All that said, the majority of our DNA doesn't appear to fall into any of these classes of "useful" sequence, so people tend to consider the rest "junk". The idea that a lot of our DNA is "junk" may be most related to the fact that about 50% of human DNA can be identified as decaying fragments of sequences that were once essentially a self-serving parasite (a virus or a "transposon"); these sequences are often called "repeats" or "repetitive sequence" because they are found everywhere in the genome and are a significant nuisance for researchers who are looking at the statistical properties of DNA sequences. If you want to know more about this, look for papers by Arian Smit, who is the author of RepeatMasker, the tool most often used to identify which "non-coding" sequence is derived from which kind of parasite; another researcher who has studied these repeats extensively is Jerzy Jurka. However, just because that sequence originated as a self-serving parasite (that didn't do the cell that contained it any good), that doesn't mean the sequence is still useless or detrimental! In fact, some of that parasite-derived sequence is now transcribed and translated into a useful protein.

So you see, the study of "non-coding" DNA is quite complex, and is alive and well. The more we learn about what our cells do with our DNA, the less of it we are likely to consider "junk". I think it would be a mistake, however, to think there is a single, amazing answer to what our non-coding DNA "does", just like it is a mistake to believe there will be a simple shot we can take to "cure" cancer or that, if we just eat one more exotic fruit per day, we'll be able to eat as many hamburgers as we want and not gain weight. Living cells are very different from the tools we design for ourselves, which tend to have a single purpose and be simple enough to understand after a short conversation. Living cells are incredibly complex, and are free to exploit any mechanism available in chemistry or physics, even at the quantum level, to perform a task that is necessary for their survival. Unlike human tools, it isn't necessary that cells work in a way that is easy for us to explain using concepts we already understand...which makes cells incredibly fascinating, but also very difficult to study.

Wow. This field has been exploding non-stop for the past decade. This is like asking if there is room for research in computers. See attached.

Non-Coding portion is overwhelming in the genome but the research in past decade was able to ascribe some function to only a minuscule of it. Still we don't know its clear role in the genome and why most of it is just sticking around the genome with no function (which could always be a possibility).

My understanding of non-coding RNAs is has crucial function in every biological pathway but we dont know well about how they are contributing to genome regulation. We just started to learn from nothing to something, but still lot to study. At least Its very clear that most of the genome is transcribing, and RNA seems to be integral part of chromatin and transcription regulation.

Yes the major part of the genomes especially human genome is non coding (~98%) but recent studies has shown that its not non-coding any more because they are coding some regulatory RNAs e.g. Non coding RNAs (micro RNA, siRNA etc) which plays key role in the development of an organism. Previously the term non-coding was assigned to the part of the DNA that don't encode any protein.

Actually these DNA are also called junk DNA as they are not expressed but recent studies show that differences between a chimpanzee and human is bcoz of these junk DNA. So probably these are involved in the evolutionary part...

Hi! Interesting topic. You guys might have seen the Nature article

http://www.nature.com/nature/journal/v478/n7370/pdf/nature10530.pdf

It gives some of the answers to you. They 'identified 3,788 candidate new exons (the genetic information for parts of proteins) with more than half of those existing OUTSIDE of known protein-coding genes' and 'loci identified in GWAS frequently lie in non-coding sequences '. Go through the paper.

There is sea of information available on noncodong RNAs. Specific roles have been assigned to specicifc types of noncoding RNAs. Finding references on this topic is no problem at all.

I think the question was about non-coding DNA, not ncRNAs. ncRNAs are very interesting and important, but the most significant function of non-coding DNA, by far, is that it contains the control sequences that are the major determinants of where and when genes will be turned on and off. The sequences are called enhancers or cis-regulatory elements; if you search those terms you will find 25 years of research. It's a very interesting field!

Yes, there was misunderstanding. You are right about role of non-coding DNA.

Lots more are there to explore. Non coding DNA part is transcribed and then spliced out and it is known that they play a major role in regulating translation. Exact mechanism is not known. They play a role also in the splicing mechanism. Can we do the knock out of noncoding part just like the mechanism is used to study the gene function?

Your question is too vague and too simple to answer in a few sentences.

"Coding" and "non-coding" are technical terms that refer specifically to the translation of messenger RNAs (mRNAs) to proteins; DNA doesn't participate in that process. The DNA bases from which the mRNA is transcribed are referred to as "coding sequence" because through transcription into RNA, they specify (or "encode") amino acid sequences. The process of translation requires RNA molecules that are also transcribed from the DNA, including transfer RNAs which hold an amino acid at one end and contact three consecutive RNA bases at the other end; these transfer RNA "genes" are part of the "non-coding" DNA, and they are needed in such quantity that there are many copies of each transfer RNA in the DNA. There are also ribosomal RNAs (rRNAs), which make up part of the ribosome structure which translates mRNAs into proteins; these genes are also non-coding genes and are present in many copies. There are several other categories of small RNAs that are transcribed from the DNA but are not translated into proteins, including microRNAs (miRNAs) which appear to help determine how often mRNAs are translated into proteins; small nuclear RNAs (snRNAs) which are required for splicing large segments (introns) out of transcribed RNAs, including mRNAs; small nucleolar RNAs (snoRNAs) which, among other things, guide the process of chemical modification of some RNAs, especially the rRNAs. Some of these, especially snoRNAs and miRNAs, are transcribed as part of introns (mentioned above) and then cut out of the intron to become functional. Even though introns seem to be useless, because they are rapidly cut out of an mRNA that goes on to be translated into a protein, it's worth keeping in mind that there may be snoRNAs or miRNAs in that intron that are also beneficial to the cell. We tend to think "linearly", i.e. one part of the DNA has one "purpose", but biology doesn't work that way. The presence of an intron is known to be important to getting a protein produced, in the sense that if you just add a gene to the DNA of human cells in culture, and that gene doesn't contain any introns, it is most likely to be transcribed and translated poorly; but inserting an intron (any intron, it doesn't appear to have to have some special component hidden in it) into that same gene often causes it to be transcribed and translated much more efficiently. So, is an intron sequence "junk" DNA, or not?

However, DNA has many functions other than encoding proteins, rRNAs, tRNAs, snRNAs, snoRNAs, and long(er) non-coding RNAs (lincRNAs)! A very important function of DNA is to be copied and delivered to the next generation, and a lot of the DNA sequence participates in that process. The centromere contains specific signals that are important in separating two copies of a DNA molecule into different cells as a cell divides. The telomeres (the ends of a linear DNA molecule, or chromosome) contain repeated signals that are bound by specific proteins to prevent the DNA from being shortened. DNA is packaged into a structure called chromatin, which requires the DNA to bend around histone proteins; this bending has been shown to be related to the DNA sequence, in the sense that there are weak, but detectable periodic patterns that result from, or contribute to, the bending process. All, or at least the vast majority, of the information that determines which genes are transcribed, and when, and in which cells, must ultimately be recorded in the DNA sequence itself; and most of that information isn't considered "coding". We have a long way to go to figure out which parts of the DNA are involved in which processes.

There are also some very puzzling regions of the genome that aren't transcribed, but they are very strongly conserved over millions of years of evolution. The fact that they are so well conserved suggests they are important for SOMETHING, but it really isn't clear yet what they do.

All that said, the majority of our DNA doesn't appear to fall into any of these classes of "useful" sequence, so people tend to consider the rest "junk". The idea that a lot of our DNA is "junk" may be most related to the fact that about 50% of human DNA can be identified as decaying fragments of sequences that were once essentially a self-serving parasite (a virus or a "transposon"); these sequences are often called "repeats" or "repetitive sequence" because they are found everywhere in the genome and are a significant nuisance for researchers who are looking at the statistical properties of DNA sequences. If you want to know more about this, look for papers by Arian Smit, who is the author of RepeatMasker, the tool most often used to identify which "non-coding" sequence is derived from which kind of parasite; another researcher who has studied these repeats extensively is Jerzy Jurka. However, just because that sequence originated as a self-serving parasite (that didn't do the cell that contained it any good), that doesn't mean the sequence is still useless or detrimental! In fact, some of that parasite-derived sequence is now transcribed and translated into a useful protein.

So you see, the study of "non-coding" DNA is quite complex, and is alive and well. The more we learn about what our cells do with our DNA, the less of it we are likely to consider "junk". I think it would be a mistake, however, to think there is a single, amazing answer to what our non-coding DNA "does", just like it is a mistake to believe there will be a simple shot we can take to "cure" cancer or that, if we just eat one more exotic fruit per day, we'll be able to eat as many hamburgers as we want and not gain weight. Living cells are very different from the tools we design for ourselves, which tend to have a single purpose and be simple enough to understand after a short conversation. Living cells are incredibly complex, and are free to exploit any mechanism available in chemistry or physics, even at the quantum level, to perform a task that is necessary for their survival. Unlike human tools, it isn't necessary that cells work in a way that is easy for us to explain using concepts we already understand...which makes cells incredibly fascinating, but also very difficult to study.

I like Max Robinson's answer.

Like this very complete reply