The 2005 signal processing review by Berryman, Allison, Wilkinson, and Abbott provides a fascinating insight into how cells operate and how similar they are to human and computer processing! 🙂 Or did all this functionality this come about by “evolution”? 🙄 Consider:
Signal processing is the use of mathematical techniques to analyze any data signal. This data could be an image, a sound, or any other sequence of data, such a sequence of nucleotides. The sequences of interest could be protein coding regions, repeating elements that may be associated with various diseases (such as Huntington’s disease) or regions rich in some set of complementary bases, such as A and T, which can give information on evolutionary history including lateral gene transfer in bacteria . . . .
Signal processing is not just a human enterprise – even individual cells process signals in the form of mRNA, protein, and more general chemical levels (for example sugars in the environment) [16, 17, 18, 19]. As with conventional computers, cells can be genetically programmed to process signals [20, 21, 22]. As in electrical circuits, switching elements can be built in, and positive and negative feedback loops are present, enabling a range of behaviours to be programmed”, such as chemical oscillations of a predetermined frequency. Such engineered \gene circuits” could have important applications in gene therapies where where we wish to modify the existing protein and cellular interactions in an organism. . . .(emphasis added)
Since DNA and amino acid sequences can be thought of as a type of language, there is interest in the use of techniques from computational linguistics to analyze genetic sequences. This theory of grammar in a computational sense was first developed by Chomsky [52, 53]. It has been applied to a wide range of applications in sequence analysis from determining gene structures  to RNA (ribonucleic acid) secondary structure . Mantegna et al. have taken methods from statistical linguistics, along with information theory approaches, to consider differences between non-coding and coding DNA [56, 57, 58]. This reveals the presence of hidden information and extra redundancy in non-coding regions, perhaps due to lengthy promoter regions , or due to information left from now defunct coding regions. A good overview of linguistic techniques used can be found in Durbin et al. .
The PROSITE database contains a large number of protein families (related sequences), and their patterns, or \motifs” [60, 61]. This database can be searched using PROSITE patterns; an example of a pattern is
where the capital letters denote amino acids. . . .
3.3. Time series analysis of gene expression data
Of interest to geneticists is not only what happens in the expression levels of two different samples at a fixed point in time, but how the expression levels vary over a number of different points in time. These experiments must be designed properly to ensure statistically signicant information can be derived . A number of different signal processing techniques have been developed to analyze such data, as well as gene clusters” (sets of related genes) in microarrays. An overview of gene clustering algorithms can be found in Moreau et al. , and below we discuss some time series approaches.
M. J. Berryman, A. Allison, C. R. Wilkinson, and D. Abbott, “Review of signal processing in genetics,” Fluctuation and Noise Letters, Vol. 5, No. 4, pp. R13-R35, 2005.
Can signal processing be explained exclusively by the four forces of nature? 😕
On Berryman et al.’s discussion of “Linguistics”, see
Prof. Dr-Ing. Werner Gitt’s hierarchy of information in his book: In the Beginning was Information ISBN: 3-89397-255-2. Gitt develops five levels of information:
- Fifth level Apobetics: Intended purpose & achieved result
- Fourth level Pragmatics: Expected and implemented actions
- Third level Semantics: Ideas communicated and understood
- Second level Syntax: Code employed and understood
- First level Statistics: Signal transmitted and received
Barryman et al’s review discuss the first level Statistics, and second level Syntax, possibly some of the 3rd level Semantics.
What basis can Darwinism provide for ANY of Gitt’s five levels of information?
We have clear current and historic evidence of intelligent agents being the direct cause of signal processing and computer programs. I find Intelligent Design provides a more satisfactory explanation for the signal processing found in cells than for the four stochastic forces of nature!
- Appendix A: Hidden Markov Models
In a hidden Markov model, the states are unknown and must be inferred from the data. . . .
- Appendix B. Fourier Transforms
Fourier transforms are used in a wide range applications such as voice prints for evidence in criminal cases, compressing images, removing noise from music, and of course in DNA sequence analysis (Subsec. 2.2) . . .
- Appendix C. Mutual Information
The mutual information function, introduced in Subsec. 2.3, for symbols at distance d apart is given in Eq. C.12, . . .
- Appendix D. Spline Curves
The use of spline curves was introduced in Subsec. 3.3. With a spline curve, one approximates a curve using a set of basic functions (often polynomials) that are fitted to the function at a set of points where the function used to approximate the curve can change, but must meet certain specications (often ones designed to make the spline curve look smooth), and conditions are also specified on the ends of the spline curve. . . .
See their signal processing review paper for all the equations etc.