Transcript Disease Informatics:Host factors simplified

Disease Informatics:
Host factors simplified
Rajendra P Deolankar
[email protected]
Laughter is the best medicine
Prerequisite
http://www.pitt.edu/~super1/lecture/lec25371/index.htm
http://www.pitt.edu/~super1/lecture/lec25381/001.htm
Modern man is genetically same as Paleolithic man
But lives in the artifact world
And hence “gene X artifact” is an important subject matter for
disease study
http://news.nationalgeographic.com/news/2004/09/images/040910_awastack.jpg
Confounders
Age, Sex, Socioeconomic status, Caste etc are not causes of
disease but could be pointers to molecules and mechanisms
causing disease
Organization of disease system
Mere set of organs is not organism
System cannot work without organization
To manage a system effectively, you might focus on the interactions of the
parts rather than their behavior taken separately. Russell L. Ackoff
Personality of disease
Disease has a personality and associated factors are its organs
Associated factors are mostly but not necessarily component
causes (CCs)
Disease Causal Mechanism (DCM)
Summarily,
Mere set of CCs  DCM
CCs: Component causes
Conceptual scheme of ageing as the accumulation of
component causes throughout life Ageing starts with the
accumulation of component causes A–E. The presence of these five
component causes completes sufficient cause I, resulting in effect I,
e.g. unsteadiness. In the following period, the addition of
component causes F–H completes sufficient cause II, resulting in
effect II, e.g. a gait disorder. The further accumulation of
component causes I and J completes sufficient cause III, resulting
in effect III, e.g. death (see also the description of the example).
http://www.biomedcentral.com/content/pdf/1471-2318-3-7.pdf
Teamwork
CCs: Component causes
P + X  PX
Where P and X are CCs and
PX is interaction / teamwork amongst CCs
PX  DCM
Disease is outcome of DCM where factors work in team
Teamwork PX predominate over individual factors P + X
DCM: Disease causal mechanism
PX: Interaction / teamwork amongst CCs
Disease and DCMs
DCM is regarded as sufficient cause
For a given disease, there could be several sufficient causes
Three sufficient causes of disease.
http://www.ajph.org/cgi/content/full/95/S1/S144/F1
Rothman and Greenland
Proposed model
PROF. Kenneth J Rothman
Prof. Sander Greenland
A model of causation that describes causes in terms of sufficient causes and their component causes
illuminates important principles such as multicausality, the dependence of the strength of component
causes on the prevalence of complementary component causes, and interaction between component
causes.
Innate α (Host X Environment)
Immunity
Specific α (Host X pathogen)
Paraspecific α (Innate X immunogen)
Disease resistance
Host is a common factor
Infectious diseases
Pathogen (P) must work together with some CC (X) to
compose DCM
CC: Component cause
P is also CC
DCM: Disease causal mechanism
What is X?
We have seen that, in an infectious diseases, Pathogen
(P) must work together with some CC (X) to compose
DCM
X could be described as complex interaction of host/
environmental factors
DCM centric Disease Definitions
DCM could be P1X OR P2X OR … PmX
Secretory Diarrhoea may be associated with E coli
toxin (P1X) OR Vibrio cholerae 01 toxin (P2X) OR
… NSP4 of rotavirus (PmX)
CC centric disease definitions
DCM could also be PX1 OR PX2 OR ... PXn
Dengue virus (P) could be associated with fever (PX1),
hemorrhagic fever (PX2) OR … Shock syndrome (PXn)
X1 ≠ X2 ≠ Xn
http://www.wordinfo.info/words/index/info/view_unit/1/?letter=B&spage=3
Crude classifications and false generalizations are the curse of organized life.
George Bernard Shaw
Disease definition for Total disease burden in a locality
DCM could be:
P1X OR P2X OR … PmX OR
P1X1 OR P1X2 OR … P1Xn OR …
P2X1 OR P2X2 OR … P2Xn OR …
………………………PmXn
Taking the wicket of a key player helps in winning the
cricket match
CC in CC centric disease definitions is regarded as a key
player
It is easy to detect CC
It is easy to plan strategy against CC
Information on DCM is complex and difficult to compile
Several captains!!! Almost no players
This could be a limitation of CC centric disease definitions
To fight Secretory Diarrhoea, solution based on E coli target
may not work against rotavirus. Ultimately several solutions
are required to fight the disease.
Team may win even after captains fails
Biologicals and chemicals prepared to fight against the disease
May fail
May produce complications
May emerge into new disease
Increase the expenditure on public health
Description of a human in Indian Medicine
Self =
Somatic body (Annamaya kosha) +
Vitality (Pranamaya kosha) +
Mind (Manomaya kosha) +
Intellect (Vidnyanamaya kosha) +
Bliss (Anandamaya kosha)
Human body computer =
Intellect (Central processing unit) + Self / Ego (Software) +
Memory (Free space, Floppy/ Hard disk) +
Mind/ senses (Program) + Life history (Data)
Principals in DCM in Ancient Indian Medicine
Errors in 4 C’s:
Catch
Control
Carry on and
Chuck
1st C
Error: In catching from nature: food, water, air, sunlight etc
Outcome: Slip in right body composition
Solution: Balancing Dosha
Dosha ≡ Body composition
2nd C
Error: In controlling the body network
Outcome: Slip in right response to the stimulus
Solution: Appropriation of Dushya
Dushya ≡ response to stimulus
3rd C
Error: In carrying on the routine
Outcome: Deviation from optimum basal metabolic rate
Solution: Regularising Agni
Agni ≡ Basal Metabolic Rate
4th C
Error: In chucking the waste
Outcome: A body that is not free from dysbiotics and morbid
substances
Prof. Stig Bengmark
Solution: Elimination of Ama
Ama ≡ products of dysbiosis
Human microbial organs
Gut associated microbiota organ
Vagina associated microbiota organ
Skin associated microbiota organ
Disease triad described in ancient Indian medicine
Disease triad is working together of
Host factors (Adhyatmic),
Environmental factors (Adhidaivik) and
Agents: physical, chemical or biological (Adhibhautik)
To give outcome as disease (Vyadhi)
Associations with disease outcome
CCs work together to give disease outcomes that can be
observed at a particular time, at a particular place or in a
particular person
Time (Season: Kala-bala),
Place (Daiva- bala) and
Person (Prakruti, described separately)
Details observed in Person
1. Genetically predisposed / metabolically imprinted (Aadibala pravritta)
2. Congenital (Janma-bala)
3. Imbalance of body composition (Dosha-bala)
4. Metabolic activity (Vata, Pitta and Kapha)
5. Trauma (Sanghata-bala) and
6. Age, sex, socioeconomic status etc (Svabhava-bala)
Thrifty genes
http://www.bmj.com/cgi/content/full/328/7447/1070
Prenatal adaptations
http://www.bmj.com/cgi/content/full/328/7447/1070
Host body phenotypic characterization
Density
Somatotype
State of matter
Composition
Motility
Shape etc
Density of body
Variation in dosha resemble density of the body
(Vata-light, Kapha-heavy; Pitta-neither heavy nor light)
Somatotype of the body
Ectomorph (Vata),
Endomorph (Kapha) and
Mesomorph (Pitta)
Mesomorph
Endomorph
http://www.innerexplorations.com/catpsy/t1c4.htm
Ectomorph
State of body matter
Gas (Vata)
Solid (Kapha) and
Liquid (Pitta)
Body composition
Low muscle (Vata)
Fatty and muscular (Kapha) and
In between i.e. lean mass dominated (Pitta)
Body motility
High (Vata)
Low (Kapha) and
Medium (Pitta)
Body shape
Linear (Vata)
Hour glass (Pitta) and
Apple, pear or rectangular (Kapha)
Tridosha in ancient medicine
Vata, Pitta and Kapha of a person are called as Tridosha in
the ancient Indian medicine
None of the body characterization criteria described in the
earlier slides singly can describe the tridosha
Tridosha is complex
Quantifying tridosha; Rajni Joshi method
http://www.liebertonline.com/doi/abs/10.1089/acm.2004.10.879?cookieSet=1&journalCode=acm
Dosha assessment
Dosha assessment may vary depending upon the skill level
of the vaidya (Doctor)
What we can learn from ancient science
Parameters for measurement of characteristics of a person
could be many and hence data could be quite huge
Dimensionality of data can be reduced if similar parameters
are grouped together
How to reduce dimensionality of data?
Techniques in multivariate statistics
Computer databases and software
Example of database preparation
Prepare a multi-dimension data set using all possible criteria
on a representative population (Density, somatotype, body
shape (digitalize), composition, IQ, EQ etc)
Example of application of multivariate statistics
Cluster the individuals by applying some algorithm
of multivariate statistics
The individuals having similar characteristic will
fall in one group
Thus the population will be divided in a few groups
(G1, G2…Gk)
Example of application of multivariate statistics
Perform Principal Component Analysis (Statistical analysis)
on each group (G1, G2…Gk)
The number of variables now would be 2 to 3.
The three components expected could be similar to Vata,
Pitta and Kapha in order of their importance
This order will vary in each group (cluster)
http://content.digitalwell.washington.edu/msr/external_release_talks_12_05_2005/13651/lecture.htm
Example of prediction model
List which variables are closely associated with Principal
Components
See how Principal Components look like in real life
How the Principal Components can be predicted?
(e.g. least square technique)
Prepare a model for component balance
Understand host with fewer parameters (Principal components)
Estimate the Prakruti (constitution of a person)
Estimate the Vikruti (Loss of harmony in constitution)
Stress and disturbed sleep are such factors which could contribute spontaneously to the DCM
The procedure described is based on
phenotypic characterization
The preventive strategies could be described
as:
Nutriphenomics
Pharmacophenomics etc
Preventive strategies
Now, human genomic data is also available
The preventive strategies are:
Nutrigenomics
Pharmacogenomics etc
Ancient Indian methods to tackle total disease burden in a
locality
Seasonal lifestyle goals (Rutucharya)
Diurnal lifestyle goals (Dinacharya)
How to implement goals
Control by risk groups (Vrata)
Transformation of patients (Vaikalya)
Festivals for everybody (Sana)
Message of Rishi Panchami
Reduce artifacts from your lifestyle
Vrata
Make up of Vrata, Vaikalya and Sana
Functional foods
Nutraceuticals
Exercises and
Spiritual practices
Dr V Prakash
Importance of Indian Medicine for Disease
Informatics
Genetic and lifestyle variables of a host could be
described in fewer words (e.g. Vata, Pitta and Kapha) to
understand events in Disease Causal Chains (DiCC)
A great help in drawing DiCC
Disease informatics for setting up Disease definition, drawing Disease Causal
Chain / Web, marking Risk Events, Backend and Frontend Events, and Health
Problem Solutions
http://bmj.bmjjournals.com/cgi/eletters/331/7516/566#134452
Thanks
Points in Indian Medicine is outcome of discussion with Dr. Mandar Akkalkotkar
Statistics guru is Dr. Sham J Amdekar