Questions on computing FI in cancer

[Patrick]

@Giltil

Entropy is information in the Shannon sense of the term. But entropy is the enemy of information in the sense of functional or specified information. This is why it is recognized that during carcinogenesis, the mutations and epigenetic alterations that occur reduce the functional information of the cell by increasing its randomness/entropy.

This is total nonsense. Entropy equals information.

[Timothy_Horton]

@Giltil

Good try, but that doesn’t work either for the information the authors are referring to is of the type that specified a function, ie, functional information.

Poor try. The papers never used the term “functional information” as you are using the term and never did any FI calculations using the formula you are using. You seem to be grasping at micron-thin straws and missing.

[Dan_Eastwood]

@Giltil

Good try, but that doesn’t work either for the information the authors are referring to is of the type that specified a function, ie, functional information.

You are prevaricating, using the term “information” to mean whatever is convenient for you at the time. That’s why you should “show your math”.

[swamidass (S. Joshua Swamidass)]

@Giltil

My claim is that carcinogenesis is a process that destroys the functional information content of cells.

That is easily disproven. We see that carcinogenesis requires cells to acquire several new and complex genetic functions.

[Giltil (Gilbert Thill)]

@swamidass

That is easily disproven. We see that carcinogenesis requires cells to acquire several new and complex genetic functions.

Maybe. But if you added up all the information produced and destroyed during carcinogenesis, the balance sheet would be largely negative. This is the message conveys not only by the two publications I offered, but also by the vast body of knowledge accumulated in oncology in the past decades.

[swamidass (S. Joshua Swamidass)]

Once again, you have to demonstrate this with more than merely your intuitions.

The fact of the matter is that new functional information is created.

[Giltil (Gilbert Thill)]

@swamidass

Once again, you have to demonstrate this with more than merely your intuitions.

I’ve given you two publications one of which stating that cancer cells tend to evolve toward a minimum information state:
The dynamics of constrained information degradation during carcinogenesis cause the tumor genome to asymptotically approach a minimum information state that is manifested clinically as dedifferentiation and unconstrained proliferation

[swamidass (S. Joshua Swamidass)]

They don’t have a calculation do they? What do they mean by information here? That isn’t evidence. It’s just an ambiguous quote from a paper.

[Giltil (Gilbert Thill)]

Their conclusion is supported by two mathematically based analysis (EPI theory and Monte Carlo experiments).

The dynamics of constrained information degradation during carcinogenesis cause the tumor genome to asymptotically approach a minimum information state that is manifested clinically as dedifferentiation and unconstrained proliferation. Extreme physical information (EPI) theory demonstrates that altered information flow from cancer cells to their environment will manifest in-vivo as power law tumor growth with an exponent of size 1.62. This prediction is based only on the assumption that tumor cells are at an absolute information minimum and are capable of “free field” growth that is, they are unconstrained by external biological parameters. The prediction agrees remarkably well with several studies demonstrating power law growth in small human breast cancers with an exponent of 1.72+/-0.24. This successful derivation of an analytic expression for cancer growth from EPI alone supports the conceptual model that carcinogenesis is a process of constrained information degradation and that malignant cells are minimum information systems. EPI theory also predicts that the estimated age of a clinically observed tumor is subject to a root-mean square error of about 30%. This is due to information loss and tissue disorganization and probably manifests as a randomly variable lag phase in the growth pattern that has been observed experimentally. This difference between tumor size and age may impose a fundamental limit on the efficacy of screening based on early detection of small tumors. Independent of the EPI analysis, Monte Carlo methods are applied to predict statistical tumor growth due to perturbed information flow from the environment into transformed cells. A “simplest” Monte Carlo model is suggested by the findings in the EPI approach that tumor growth arises out of a minimally complex mechanism. The outputs of large numbers of simulations show that (a) about 40% of the populations do not survive the first two-generations due to mutations in critical gene segments; but (b) those that do survive will experience power law growth identical to the predicted rate obtained from the independent EPI approach. The agreement between these two very different approaches to the problem strongly supports the idea that tumor cells regress to a state of minimum information during carcinogenesis, and that information dynamics are integrally related to tumor development and growth.

[Roy (34-year veteran)]

@Giltil

@Dan_Eastwood

There must be many ways to mutate a normal lung cell and still have a perfectly functional lung cell

Still, there are much more ways for a cell to be a cancer lung cell than to be a normal lung cell.

You keep saying that - but you have given no reason for anyone to accept your word for it.

This is because mostly all the mutations that could occur in a normal lung cell without altering its lung function wouldn’t affect the function of the cancer lung cell either. But the opposite is not true. Indeed, many mutations that could occur in a cancer lung cell without altering the cancer function would affect the function of the normal lung cell.

…at which point it would not be a cancer lung cell - so it wouldn’t be relevant to your calculation.

[Roy (34-year veteran)]

@Giltil

Notice also that the idea that carcinogenesis is a process that destroys the functional information of cells is basic knowledge in oncology.

I would be extremely surprised if you could cite any standard oncology work that even mentioned “functional information”.

By denying this well established fact, you’re betting on the wrong horse

You saying something does not make it well established, no matter how many times you repeat yourself.

[Roy (34-year veteran)]

@Giltil

My claim is that carcinogenesis is a process that destroys the functional information content of cells. The two publications I gave support this claim. Period.

Neither abstract mentions “functional information”. Can you cite the passages from those works that do mention it, or should I simply conclude that you’re lying?

[Dan_Eastwood]

I just finished reading Gatenby (2004). To very briefly summarize, some information is lost or transformed to promote cancer cell fitness, while other information is maintained or increased to keep the cell alive. This is just as SJS described using Venn Diagrams in the progenitor thread.

There is no mention of “Functional Information” in this article, only standard Information Theory measures: Shannon, Fisher, mutual, relative entropy (Kullback–Leibler divergence).

[Giltil (Gilbert Thill)]

@Dan_Eastwood

I just finished reading Gatenby (2004). To very briefly summarize, some information is lost or transformed to promote cancer cell fitness, while other information is maintained or increased to keep the cell alive.

Your summary is misleading for it gives the impression that losses and gains of information are in balance when in reality the losses are prominent to the point that tumor cells tend to approach a minimum information state.

The dynamics of constrained information degradation during carcinogenesis cause the tumor genome to asymptotically approach a minimum information state that is manifested clinically as dedifferentiation and unconstrained proliferation

@Dan_Eastwood

There is no mention of “Functional Information” in this article, only standard Information Theory measures: Shannon, Fisher, mutual, relative entropy (Kullback–Leibler divergence).

Look. Consider the two sequences of letters below:

1. ehovcssfpmkjhtijfujfazembhdohfdqtplnvc
2. geneticinformationisfunctionalinformation

These two sequences have about the same amount of Shannon information. And yet, they are very different. Do you see why?

[Giltil (Gilbert Thill)]

@Roy

I would be extremely surprised if you could cite any standard oncology work that even mentioned “functional information”.

Standard oncology work mention degradation of genetic information in cancer cells. But since genetic information specify functions (ie, is functional), it follows that saying that there is degradation of genetic information in cancer cells is equivalent to saying that there is degradation of functional information in cancer cells.

@Roy

This is because mostly all the mutations that could occur in a normal lung cell without altering its lung function wouldn’t affect the function of the cancer lung cell either. But the opposite is not true. Indeed, many mutations that could occur in a cancer lung cell without altering the cancer function would affect the function of the normal lung cell.

…at which point it would not be a cancer lung cell

???!

[Dan_Eastwood]

@Giltil

@Dan_Eastwood

I just finished reading Gatenby (2004). To very briefly summarize, some information is lost or transformed to promote cancer cell fitness, while other information is maintained or increased to keep the cell alive.

Your summary is misleading for it gives the impression that losses and gains of information are in balance when in reality the losses are prominent to the point that tumor cells tend to approach a minimum information state.

The dynamics of constrained information degradation during ca

I am accurately providing additional information from the article which you have omitted.

@Giltil

The dynamics of constrained information degradation during carcinogenesis cause the tumor genome to asymptotically approach a minimum information state that is manifested clinically as dedifferentiation and unconstrained proliferation

You can stop quoting the abstract at me, I’ve read the whole paper.

@Giltil

@Dan_Eastwood

There is no mention of “Functional Information” in this article, only standard Information Theory measures: Shannon, Fisher, mutual, relative entropy (Kullback–Leibler divergence).

Look. Consider the two sequences of letters below:

1. ehovcssfpmkjhtijfujfazembhdohfdqtplnvc
2. geneticinformationisfunctionalinformation

These two sequences have about the same amount of Shannon information. And yet, they are very different. Do you see why?

Thank you, I better understand what you are trying to say now. However, you are mistaken. Shannon Information does not encode meaning, only the variability of the source population. The “minimum information state” refers to the mutual information shared between the original lung cell line (the sender) and the evolved cancer cell (the receiver). The physical interpretation is that only part of the original DNA is preserved by the cancer cell, and the function of being a “lung cell” has been lost. Total Shannon Information remains about the same (assuming no insertions or deletions). The cancer cell has gained some new Information too (the driver mutations SJS mentioned and more) which enable it’s new function of being a cancer cell.

So of course a cancer cell is different from the original lung cell, no one is arguing that. It may be true that a cancer cell needs less DNA to encode its new function than the original lung cell function, and is “simpler” in this sense. But these are different functions, making it an apples and oranges comparison. Something like this, where row zero is an attempt to label parts of the message:

0. MUTUAL_UNFORMATIONnoiseNEW_FUNCTIONnoise 
1. geneticinformationlnvcrtheskyisbluekjqtp4
2. geneticinformationisfunctionalinformation

Two sentences with different meanings or functions. We would not say one sentence is less meaningful or less functional than the other because it is shorter. The length of the message is irrelevant to meaning or function.

[Giltil (Gilbert Thill)]

@Dan_Eastwood

The “minimum information state” refers to the mutual information shared between the original lung cell line (the sender) and the evolved cancer cell (the receiver). The physical interpretation is that only part of the original DNA is preserved by the cancer cell, and the function of being a “lung cell” has been lost. Total Shannon Information remains about the same (assuming no insertions or deletions). The cancer cell has gained some new Information too (the driver mutations SJS mentioned and more) which enable it’s new function of being a cancer cell.

I agree with nearly all you are saying here except for the fact that the minimum information state (MIS) of a cancer cell refers to the MI. Note that if MIS=MI, then you would have to recognize that the FI of a cancer cell is smaller than the FI of its non cancerous counterpart.

@Dan_Eastwood

The length of the message is irrelevant to meaning or function.

Yet, the length of a message is relevant to its FI content.

[Roy (34-year veteran)]

@Giltil

@Roy

I would be extremely surprised if you could cite any standard oncology work that even mentioned “functional information”.

Standard oncology work mention degradation of genetic information in cancer cells. But since genetic information specify functions (ie, is functional), it follows that saying that there is degradation of genetic information in cancer cells is equivalent to saying that there is degradation of functional information in cancer cells.

As expected, no cites at all.

[Dan_Eastwood]

@Giltil

I agree with nearly all you are saying here except for the fact that the minimum information state (MIS) of a cancer cell refers to the MI. Note that if MIS=MI, then you would have to recognize that the FI of a cancer cell is smaller than the FI of its non cancerous counterpart.

Note: We should be care of terms. I assume that you mean MI=Mutual Information? In all cases here I am referring to Shannon Information and related measures.

Mutual Information will be less than the MIS (minimum info state), because MIS will include any new information in the cancer cell.

Shannon Information does not encode messages. It is a measure of information carrying capacity. More bits means greater capacity, but meaning (and function) depend on the encoding/decoding of the sender/receiver. Further, the coding need not be “efficient”, so the encoding need not use all bits available (ie: less Information). Short messages can be very meaningful/functional, and vice-versa.

BUT

Shannon Information does not capture meaning or function, and it is not used in that sense in the paper. Whatever it is you mean to say by “Functional Information”, either you are using the wrong term, misstating, or it’s simply not there.

Have you actually read anything more than the abstract?