Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influenza

[AlaskaDenise]

1: Amino Acids. 2007 Nov 2; [Epub ahead of print]

Wu G, Yan S.
Computational Mutation Project, DreamSciTech Consulting, Shenzhen, Guangdong Province, China.

This is the continuation of our studies on the prediction of mutation engineered by randomness in proteins from influenza A virus. In our previous studies, we have demonstrated that randomness plays a role in engineering mutations because the measures of randomness in protein are different before and after mutations. Thus we built a cause-mutation relationship to count the mutation engineered by randomness, and conducted several concept-initiated studies to predict the mutations in proteins from influenza A virus, which demonstrated the possibility of prediction of mutations along this line of thought. On the other hand, these concept-initiated studies indicate the directions forwards the enhancement of predictability, of which we need to use the neural network instead of logistic regression that was used in those concept-initiated studies to enhance the predictability. In this proof-of-concept study, we attempt to apply the neural network to modeling the cause-mutation relationship to predict the possible mutation positions, and then we use the amino acid mutating probability to predict the would-be-mutated amino acids at predicted positions. The results confirm the possibility of use of internal cause-mutation relationship with neural network model to predict the mutation positions and use of amino acid mutating probability to predict the would-be-mutated amino acids.

PMID: 17973072 [PubMed - as supplied by publisher]

http://www.ncbi.nlm.nih.gov/sites/entrez?tmpl=NoSidebarfile&db=PubMed&cmd=Retrieve&list_uids=17973072&dopt=AbstractPlus

Hopefully "someone" can translate!

.

[cartski]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

"Hopefully "someone" can translate! "

Ditto!!!

What does this study get at? What does it mean? What does it say about the direction and quality of Chinese research in this area?

J.

[AlaskaDenise]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Some of the statements seem contradictory to me. It looks like it could be a study to support a preconceived concept, while ignoring other possibilities.

.

[wetDirt]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Here is something like what I think they are doing:

http://www.msm.cam.ac.uk/phase-trans/2006/ga_html_files/ga.html

Basically, they isolate rules like restrictions on the location of a 'mutation' and its 'fitness', but including more than just these factors. Perhaps they have a rule like, "i before e except after c" that makes some kinds of mutations impossible. They then assign random weights and calculate all the possible combinations, then weed out the impossible ones, then figure out what rules turn out to be the essential ones.

Once they know what combinations of factors turn out to be the most important, then they can use that to predict, within limits, what the system will do.

This kind of analysis doesn't necessarily require that the mutations be random, only that the reasons for the mutations are obscure and the relevant drivers aren't known.

Do I need to add that if you don't include enough of the relevant factors, you won't get a reliable model?

[AlaskaDenise]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Do I need to add that if you don't include enough of the relevant factors, you won't get a reliable model?

i.e., the whole unexplored field of epigenetics?


I'm less inclined to believe influenza operates on any mathematical model and more inclined to believe it responds to situations (unless the situations are the rules they are evaluating).

If the genetic sequences database is only about 100 years old, that may not be enough history to arrive at reliable conclusions. For instance, since man domesticated ducks about 4500 years ago, how can we evaluate that influence without a "before and after" data set?

.

[wetDirt]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Well, yes, in fact. But (without, of course, having actually read the article) I am thinking they will mainly discover that the physical shape of the surface proteins determines how will the 'key' will fit the 'lock', and that some mutations ruin the fit because they put a bend where it shouldn't be. So they are going to find out that there are only a few ways to make a key, that bad fits are rapidly eliminated, and that it's easier to keep a big pool of good subunits floating around than to constantly rewrite Shakespeare using monkeys. Ummmmm, but they will show it nicely with an expert system.

[DreamSciTech]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Dear All

It is by chance that we found that our recent paper is being discussed in this forum when we were searching through Yahoo.

We as authors ourselves are very happy to answer any question regarding this paper. From the discussion posted in this site, we would like to explain a little.

We use a cause-mutation relationship to predict the mutation, because the best way to predict the mutation is perhaps to find the cause for each mutation, as we could predict a mutation when its cause appears.

However, many causes, which led to historic mutations, left no trace because of huge changes in environments, thus we might have a relatively detailed record of mutations, but a poor record of their causes. Second, the current version of proteins may no longer be subject to the causes, which led to mutations in the past, because of evolution of proteins. Third, we have a great difficulty to find the microenvironment, under which the historic mutation causes functioned.

Therefore we consider using the randomness as a consistent cause engineering mutations, by which we build a cause-mutation relationship by coupling quantified randomness as cause and occurrence/non-occurrence of mutations as mutation to make the prediction.

We welcome any your questions on our approach, and we would like to invite you to visit our website, www.dreamscitech.com, which we just reconstructed, including our monitoring of evolution of hemagglutinins using our methods with the hope of predicting of mutations.

Regards

Guang Wu
Shaomin Yan
DreamSciTech

[sharon sanders]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Welcome Guang and Shaomin.

I have a question - a theoretical one.

If we can predict randomness then is it random? And if it is, then isn't it random because we do not yet know all the variables?

[DreamSciTech]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

In prediction of mutation using mathematic tools, we must have an equation, say, ax + b = y, if y is the mutation, x must be something that have a value. We defined x with quantified randomness in order that x has some values.

[sharon sanders]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

From your website:

"...The pattern in this figure highlights that a pandemic/epidemic generally follows the extreme point of quantified hemagglutinins....As can be seen in this figure, our current position is approaching to an extreme."

Interesting......

[DreamSciTech]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Thanks!

In figure in www.dreamscitech.com, you can see the importance of quantifying a protein, that is, we need to use a single value to present a whole protein. This value should be subject to mutation, by such a value we can follow the historic path of hemagglutinins through a 2-dimensional graph, then we can approximately make prediction by pattern comparison, which we may call the dynamic approach.

[AlaskaDenise]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Welcome Guang and Shaomin.

We appreciate you taking your valuable time to explain your research.

Your work sounds fascinating & should be a valuable contribution to the field.

It sounds as if it will provide valuable support the work of epigeneticists.

.

[Sally Furniss]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Neural Networks Versus Logistic Regression In Financial Modelling: A Methodological Comparison

Abstract: this paper is to very briefly examine two alternative methodologies which might be applied to decision making classification problems in corporate finance: logistic regression and neural networks. To achieve this end, the basic concepts underpinning each method are described. The nature of modelling the corporate financial decision making environment is then discussed. Suggested methodologies are then prescribed to aid those interested in using either or both techniques, and salient results...

http://citeseer.ist.psu.edu/73860.html

[Laidback Al]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Guang and Shaomin, thanks for taking the time to join FluTrackers and answer some questions about your recent article. Most of us here at FluTrackers do not have full access to journal articles, so I was wondering if you would briefly discuss your results regarding the timing of influenza outbreaks based on the research you published in 2005. Abstract:

Timing of mutation in hemagglutinins from influenza A virus by means of unpredictable portion of amino-acid pair and fast Fourier transform

Guang Wu, and Shaomin Yan
Computational Mutation Project, DreamSciTech Consulting, 301, Building 12, Nanyou A-zone, Jiannan Road, Shenzhen, Guangdong Province CN-518054, China
Received 22 April 2005. Available online 26 May 2005.


Abstract

In this study, we calculate the unpredictable portion of amino-acid pairs, which has been developed by us over the last several years, of 1201 hemagglutinins from influenza A viruses dated from 1918 to 2004 in order to compare them with respect to subtypes, species, and years. After noticing the fluctuations of unpredictable portion along the time course, we use the fast Fourier transform to find the mutation periodicity of hemagglutinins. Then we estimate our position at the current cycle of hemagglutinin evolutionary process to determine how many years remain before the next outbreak of influenza and bird flu. Finally, we use the trend line and channel to outlook the hemagglutinins for the next half a century. As our study covers almost all the full-length amino-acid sequences of hemagglutinins from various influenza A viruses, the conclusion will be valid for years until the number of hemagglutinins in protein databank will be significantly increased.


Keywords: Fast Fourier transform; Hemagglutinin; Influenza A virus; Mutation; Periodicity



Biochemical and Biophysical Research Communications
Volume 333, Issue 1, 22 July 2005, Pages 70-78

[AlaskaDenise]

Re: Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influe

Fast Fourier transform

From Wikipedia, the free encyclopedia


Jump to: navigation, search
<!-- start content -->"FFT" redirects here. For other uses, see FFT (disambiguation).

A fast Fourier transform (FFT) is an efficient algorithm to compute the discrete Fourier transform (DFT) and its inverse. FFTs are of great importance to a wide variety of applications, from digital signal processing and solving partial differential equations to algorithms for quick multiplication of large integers. This article describes the algorithms, of which there are many; see discrete Fourier transform for properties and applications of the transform.

Let x<SUB>0</SUB>, ...., x<SUB>N-1</SUB> be complex numbers. The DFT is defined by the formula
<DL><DD></DD></DL>Evaluating these sums directly would take O(N<SUP>2</SUP>) arithmetical operations. An FFT is an algorithm to compute the same result in only O(N log N) operations. In general, such algorithms depend upon the factorization of N, but (contrary to popular misconception) there are FFTs with O(N log N) complexity for all N, even for prime N.

Many FFT algorithms only depend on the fact that is a primitive root of unity, and thus can be applied to analogous transforms over any finite field, such as number-theoretic transforms.

Since the inverse DFT is the same as the DFT, but with the opposite sign in the exponent and a 1/N factor, any FFT algorithm can easily be adapted for it as well.

<TABLE class=toc id=toc summary=Contents><TBODY><TR><TD>Contents

[hide]

• <LI class=toclevel-1>1 The Cooley-Tukey algorithm <LI class=toclevel-1>2 Other FFT algorithms <LI class=toclevel-1>3 FFT algorithms specialized for real and/or symmetric data <LI class=toclevel-1>4 Bounds on complexity and operation counts <LI class=toclevel-1>5 Accuracy and approximations <LI class=toclevel-1>6 Multidimensional FFT algorithms <LI class=toclevel-1>7 Other generalizations <LI class=toclevel-1>8 See also <LI class=toclevel-1>9 References
• 10 External links

</TD></TR></TBODY></TABLE><SCRIPT type=text/javascript>//<![CDATA[ if (window.showTocToggle) { var tocShowText = "show"; var tocHideText = "hide"; showTocToggle(); } //]]></SCRIPT>
The Cooley-Tukey algorithm

Main article: Cooley-Tukey FFT algorithm.
By far the most common FFT is the Cooley-Tukey algorithm. This is a divide and conquer algorithm that recursively breaks down a DFT of any composite size N = N<SUB>1</SUB>N<SUB>2</SUB> into many smaller DFTs of sizes N<SUB>1</SUB> and N<SUB>2</SUB>, along with O(N) multiplications by complex roots of unity traditionally called twiddle factors (after Gentleman and Sande, 1966).

This method (and the general idea of an FFT) was popularized by a publication of J. W. Cooley and J. W. Tukey in 1965, but it was later discovered that those two authors had independently re-invented an algorithm known to Carl Friedrich Gauss around 1805 (and subsequently rediscovered several times in limited forms).

The most well-known use of the Cooley-Tukey algorithm is to divide the transform into two pieces of size N / 2 at each step, and is therefore limited to power-of-two sizes, but any factorization can be used in general (as was known to both Gauss and Cooley/Tukey). These are called the radix-2 and mixed-radix cases, respectively (and other variants such as the split-radix FFT have their own names as well). Although the basic idea is recursive, most traditional implementations rearrange the algorithm to avoid explicit recursion. Also, because the Cooley-Tukey algorithm breaks the DFT into smaller DFTs, it can be combined arbitrarily with any other algorithm for the DFT, such as those described below.


(The article continues....)