Factoring Out Common Genes To Find Unknown Ones

ProgressiveCynic writes "Dr. Michael Brent has designed a novel algorithm for discovering unknown genes by factoring out commonalities between the genomes of different species. Using the algorithm to compare the genomes of mice and men, he and his collegues have already discovered over a thousand new genes.
So why didn't he just use ZIP?"

Facorting out Common Stories To Find New Ones

[zulux]

Dr. Michael Brent has designed a novel algorithm for discovering new Slashdot stories by factoring out commonalities between the sotries of diferent days. Using the algorithm to compare the sories of today and of a few days ago, he and his collegues have already discovered over a thousand Slashdot stories that haven't even been noticed.

Re:Facorting out Common Stories To Find New Ones

[umofomia]

Dr. Michael Brent has designed a novel algorithm for discovering new Slashdot stories by factoring out commonalities between the sotries of diferent days. Using the algorithm to compare the sories of today and of a few days ago, he and his collegues have already discovered over a thousand Slashdot stories that haven't even been noticed.

On an interesting sidenote, during his analysis Dr. Brent also discovered that many of the stories appeared to be duplicates of each other. The evolutionary purpose and reason for these duplicates are still under speculation, but he has been overheard mentioning the word "CmdrTaco".

Clothes shopping

[Bisifiniti]

I often factor out common jeans to find a really good pair.

Did the author really just say that?

[immanis]

Looks to me like the author just told Dr. Michael Brent to Zip his Genes.

Did he leave the barn door open?

Of Mice and Men

[74Carlton]

There are between 25,000 and 30,000 genes in both the human and mouse genomes, with no more than 500 genes separating the two mammals.

No wonder I like granola.

Huh?

[cornjchob]

He factored out genes from Mice and Men? I take it latter research will lead to factoring out the genes of the Lord of the Flies.

Re:Huh?

[MrLint]

Tell me about the genes of the rabbits george!

"factoring out?"

[MarkusQ]

It sounds like what they are doing is identifying the shared patterns (rather than "factoring them out"); the idea being that, if a gene codes for something important it won't change much (important gene + random change = dead critter), while the junk regions will change at a steady rate. If you look at two creatures that are medium-distant relatives, the sequences they share are likely to be important genes.

This is exactly oposite the usual connotation of "factoring out," which implies that they are removing the common regions and looking for the differences.

-- MarkusQ

P.S. And, with regard to ZIP, they aren't looking for repetitions within a genome, but rather similar regions in corresponding portions of different genomes.

Yawn

[goombah99]

Umm...I wont critisize Michael brent, since no doubt he is the victium here of the usual confused tech reporting and hyper-press release. But the author of the article does not seem to appreciate that this is NOTHING NEW. this is what EVERYONE in the buisness does. 60% of all genese in every genome are annotated by sequence similarity to other genomes. using evolutionary distance and sequence substitution is what EVERYBODY does. there almost isn't another way. Indeed that's where all the researchin the field lies, figuring out another way besides sequences imilarity to recognize genes.

So what did he doe that was original. Well We'll never know from the article since its content free. One could speculate from the hints in the article that hes is concentrating on gene finding and not gene annotation. the difference is that in eucariots as opposed to bacteria, its harder to actually spot genes in the DNA. thus any trick you can to hep you spot genes is helpful. but it's not earthshaking.

yawn

Biological Significance?

[spotted_dolphin]

Although it appears valid how Dr. Brent is attempting to 'uncover' these novel genes, I'm not sure how much of a breakthrough this is. We have the sequences of both species and it was inevitable that an algorithm would eventually have been developed to hunt for them. There is nothing to be said about the biological relevance of his findings. Scientists will still have to conduct many experiments to determine if any of these novel genes are truly expressed let alone what the function of the gene products are, if any. Plus, I'm not sure what he means when he says that 7000 experiments will be reduced to 1500. I don't believe he has had much experience in a molecular biology lab. These findings will reduce the amount of effort biologists will exert to find the novel genes, but there is still a lot of work yet to be completed.

Re:Biological Significance?

[gene_tailor]

Yes, you've put your finger on it-- as the scientists say in their PNAS abstract [pnas.org], one of the points of doing mouse sequence was to use it as a basis for comparisons like this, so this is interesting but certainly not big news.

As for the molecular biology aspects, the abstract states that they verified the existence of "112 previously unknown homologs of known proteins" , at least in some cases using RT_PCR (which suggests that the gene is real since it is expressed as mRNA). Apparently the "1019" refers to 907 additional predictions for which they have only computer-based speculations. Clearly it's nice to have some predictions to work from, but that's all this is.... wish I could supply more info but I can't get to the full text article without a subscription.

Newsworthy?

[devbiowonk]

This is an interesting paper, but I have no idea why it made it into the mainstream press. The concept of using comparative genomics to identify ORFs is not novel; however, the results of using this tecnique on the mouse a human genomes are interesting. Now that we have the genomes of all of these organisms(mouse, human, fly, worm, fish coming soon) the largest obstacle is just sorting out the genes from all of the crap dna accumulated over the years. These researchers found a large group of genes that hadn't been previously identified by conventional methods designed for analyzing just one genome. They consequently confirmed their data by isolating the RNA from these genes by RT-PCR showing that they were at least transcribed. The highlight of the paper is the apparent lack of conservation beween the mammalian genomes and the fish(I think that they used fugu)as two-thirds of their "new" genes don't have homologues in the fish. Why this paper was in the news-I have no idea. Besides, the article was in PNAS!