Google Tech Talk by Andras J. Pellionisz
October 30, 2008

See YouTube here




In 18 months full human genome sequences will be available under $100 - and in minutes. The $5,000 full human genome was announced to come in 9 months. Is "Big IT" ready for the avalanche of data, to be obtained and processed e.g. while the patient is still on the operating table, to be diagnosed, and how the genomics glitch, that caused a benign or malignant tumor, could be compensated for?

Algorithmic approaches are needed to better understand genome regulation, even for the simple reason to deploy most effective data retrieval, data storage and computational means, via both parallel hardware and software, but more importantly for opening entirely new perspectives.

In the 100+ year old Genomics, for over half a Century had us to resign to the fatalistic gloom that we are stuck with any glitches in our inherited genome. Is it true that genomic glitches doom one to "incurable" hereditary diseases?

No longer. Genomics now considers the DNA-RNA-Protein chain not as a thermodynamically closed system, where entropy increases, but as an open system that can be interfered with. There is theoretically sound hope that you are not stuck with your genomic glitches.

After half a Century of sticking to two mistaken axioms of Genomics, the paradigm of recursive genome function must quickly make up for lost time for those (potentially) inflicted with formerly "incurable" diseases. "The Genome baby is left on the doorsteps of Information Technology".


For skipping to section of interest, scroll down on still-pictures and note the time-stamp e.g. 27:49/57:59 for "The Principle of Recursive Genome Function". After launching YouTube in the frame on the top of this page, slide the toggle-switch on the bottom of the video-page to the desired time-stamp.

Click and slide toggle to 01:27

Hologenomics is Genomics and Epigenomics expressed in Informatics. HoloGenomics Society, HolGenTech, Inc.

Click and slide toggle to 02:28

Genetics, EpiGenetics, Genomics, EpiGenomics, PostGenetics integrated into the holistic approach to heredity; HoloGenomics

Click and slide toggle to 04:35

Genomics became Informatics (neutral declaration by Richard Dawkins, 2008). Heredity looked from the viewpoint of Information Theory appeared to be a tiny Genome Baby (of 40Mb size)

Click and slide toggle to 06:27

The Genome Baby was left on the doorstep of Informatics

Click and slide toggle to 06:56

Sergey Brin's investment into Anne Wojcicki's 23andMe picks up 1.6M Single Nucleotide Polymorphisms to establish the Personal Genomics "Direct to Customers" business model.

Click and slide toggle to 07:55

Sergey Brin's LRRK2 shows SNP, elevating risk of Parkinson's Disease. The Principle of Recursive Genome Function re-defines risks of genomic glitches.

Click and slide toggle to 11:42

Genome Informatics is segmented by Pellionisz to "Get info" (sequencing) and "Use info" (downstream sequence analysis). Both sectors are extremely computation-intensive; with diagnosis and biodefense speed-critical.

Click and slide toggle to 12:38

For computing in Genome Informatics speed is of the essence.

Click and slide toggle to 12:54

Big Pharma is a proven huge market for Genome Informatics. Merck bought Small Interfering RNA-mining company ("use info") for $1.1. Bn in 2006

Click and slide toggle to 13:33

Big IT is a proven huge market for Genome Informatics. Intel invested $100 M for 3rdGen (Single Molecule Nanosequencing of DNA) by Pacific Biosystems. Big IT profits from both "get info" (sequencing) and "use info" (diagnosis based on inexpensive full human sequences in 9-18 months)

Click and slide toggle to 14:26

Singularity of inversion of cost of sequencing and value of diagnosis, prevention and eventually a genome-based economy creates Genome Computer (GC) Era - Bigger than PC Era

Click and slide toggle to 17:21

Genome Informatics data centers could be modeled after Google, may follow cluster solutions or parallel (hybrid) computing

Click and slide toggle to18:21

DRC in leads a Xilinx-AMD-HP approach to hybrid (serial and FPGA) architecture, a suitable prototype of Genome Computer hardware.

Click and slide toggle to 19:30

XtremeData covers the market by offering both Altera/AMD and Altera/Intel "in socket" hybrids, suitable as Genome Computer, integrated e.g. by HP.

Click and slide toggle to 21:06

Genome Computer hybrid hardware can be leveraged by at least 16 combinations of Xeon/Opteron, Xilinx/Altera, HP, IBM, Dell, Apple (etc). Key is Genome Informatics Operating System and Killer Applications software, requiring deep domain expertise both in computing and genomics.

Click and slide toggle to 22:50

Genome Information Technology can only be built, if the necessary Genome Information Theory is created. ENCODE in 2007 revealed that long-held beliefs were needed to be re-thought.

Click and slide toggle to 23:44

In 2007, Genome Information exploded over 25 orders of magnitude. Genome regulation (08 March Science cover by Mattick) shows that information-theoretical assembly of knowledge is required, starting with reversal of two mistaken dogmas.

Click and slide toggle to 25:51

The "Central Dogma of Molecular Biology" (by Crick, since 1956) is false. (See analysis in science paper)

Click and slide toggle to 27:49

Principle of Recursive Genome Function by Pellionisz (submited 2007) superseeds dogma by reversal of two mistaken axioms (Central Dogma and Junk DNA)

Click and slide toggle to 29:19

Gene paradigm of forward growth ruled for half a Century; DNA>RNA>PROTEINS

Click and slide toggle to 30:35

FractoGene: fractal DNA governs recursive iterative growth of protein structures, with perused auxiliary information packets (formerly, junk) cancelled by methylation.

Click and slide toggle to 32:55

Theoretical significance of the Principle of Recursive Genome Function is, for physicists, that the "Entropy increases in closed system" does not apply to the open system of the HoloGenome. Intrinsic (gene-induced) and extrinsic proteins (e.g. prions) interact with the DNA>RNA>PROTEIN>DNA.... etc recursion.

Click and slide toggle to 34:43

Significance of "The Circle of Hope" of recursion is, that hereditary diseases are no longer a priori "incurable".

Click and slide toggle to 36:15

Venture philanthropists like Eli Broad, Sergey Brin, Brad Margus, Marty Tenenbaum (etc) are cardinal to inspiration and support of the fight against no-longer-in-principle-incurable hereditary diseases

Click and slide toggle to 36:37

Fractal approach by Pellionisz is at the stage of deployable theory, algorithms and software to Genome Computers to identify Fractal Defects, disrupting key regulatory mechanisms, thus already linked to scores of hereditary diseases. E.g. tandem repeat disrupted by binucleotide run is shown in SNCA promoter, associated with Parkinson's disease. ~3,000 structural aberrations are catalogued for PD and Alzheimer's. Which is harmless and which is a fractal defect in a key regulatory sequence?

Click and slide toggle to 38:18

Once genomic glitches are there, they are no longer in principle incurable. Epigenetic breakthrough by Waterland et al. shows that in certain cases preventive measures as simple as proper diet (tailored to the personal genome) eliminate or alleviate ill-effects.

Click and slide toggle to 39:43

Chinese empirical medicine suggested for Millennia (and now verified by science papers) that green tea may be beneficial against many diseases including cancer, cardiovascular disease and Parkinson's disease)

Click and slide toggle to 40:19

Stages are 1) Diagnosis for patients, 2) Prevention for health-partners, 3) Personal Genomic guidance to customers, towards 4) Genome-driven Economy: Consumer choices to Fit or Fix you Genome

Click and slide toggle to 41:39

System-design from USPTO submission (Pellionisz) for Genome-Driven economy. Barcode-reader PDA, synced with Genome Computer matches personal genome with best fitting items (foods, drugs, environment, etc).

Click and slide toggle to 43:19

For Google-audience, recursive iteration and fractality are not only concepts that are immediately grasped, but 10,000+ programmers are up and running for application of software-enabling HoloGenomics. Old School Genetics (after its first Century) needs time to accomodate Informatics.

Click and slide toggle to 43:49

Question and Answer Section of Pellionisz' Google Tech Talk, ending at 57:59. With further questions or requesting information contact


A peak of publicity was created, since "2008 is the Year when Genomics became Informatics". Intel's $100 M investment into Pacific Bioscience (3rd Generation Molecular Sequencing startup) is, unfortunately though somewhat predictably, perceived as "turf war" if after its first 100 years Genomics is no longer the monopoly of Biochemistry, but Informatics "intrudes".

Stonework of Egyptian Hieroglyphs, however, could be forever studied if the tables were made of marble, granite, sandstone, limestone - whatever, the material itself would never have explained the picture-coded message. Nor would any "full sequencing" of A,C,T,G ever become "readable" by itself. Of course, crystallographers can be disappointed, even to the degree of "sour grapes" when it becomes obvious that the job is up to cryptography-decoders, rather. Some "crystallographers" may even resort to badmouthing those in search of the meaning rather than letters, or the material of the letters. Sour-graping is not very helpful - but emotions are often revealing*.

After two weeks, Google yields 14,400 hits for "Is IT ready for the Dreaded DNA Data Deluge"
Blogs discussing the presentation are here here here here here here
Some relevant websites are here

* Footnote