Individuals respond differently to administration of drugs, exposures to xenobiotics, and in general to many other epidimiological and environmental factors. In the postgenomic era understanding is coming that to a major extent individual succeptibility and resistance are genetically predetermined. They are associated with polymorphic variations in genes, microDNA, and other functionally important regions of the genome. How these pieces co-express together making-up each individual phenotype, perhaps, is the next greatest puzzle after the genomic text itself.

is a web-service that models a population distribution of health effects outcomes that are associated with polymophic variations in multiple genes. It combines together in one joint distribution idiosyncratic gene / health-effect associations expressed by Odds Ratios (ORs), and estimates the probability of each possible genotypic combination from the observed idiosyncratic (marginal) gene frequencies. Next, a distribution of cumulative risk attributed to polygenic variations is determined. When citing POLYMARK in the literature, the following citation should be used:

Demchuk, E., Yucesoy, B., Johnson, V.J., Weston, A., Germolec, D.R., De Rosa, C.T. & Luster, M.I. (2007) A statistical model to assess genetic susceptibility as a risk factor in multifactorial diseases: Lessons from occupational asthma -- Environ. Health Persp. 115(2), 231-234.

The form accepts idiosyncraitic gene data. Generally, the more genes the better, however, it takes noticable amount of time to calculate more than 20 genes/biomarkers. During that time your browser may or may not time out. Check out the current version limits below. The input shall be tabulated in three columns, one marker per row. The 1^st column is an odds ratio reported in epidimiological or clinical trials, the 2^nd is the frequency of genetic biomarker associated with the health effects in controls, and the 3^rd is the frequency of biomarker in cases. Follow the example below, or you may copy it to the form to test:

2.46	0.117021276596	0.245551601423
2.24	0.033707865169	0.072386058981
1.50	0.223443223443	0.302127659574
2.70	0.089201877934	0.209302325581
7.76	0.018691588785	0.128712871287
3.14	0.098360655738	0.255319148936
1.83	0.108651911469	0.182352941176
0.28	0.289473684211	0.101694915254
2.23	0.218649517685	0.384615384615
8.18	0.017543859649	0.127516778523
4.47	0.160000000000	0.460000000000

where the columns are

< Odds Ratio >

< frequency in CONTROLS >

< frequency in CASES >

At present the online version depends on the following approximations and limitations:

• genes/biomarkers are mutualy independent
• not more than 23 genes/biomarkers can be processed at once

________

Saturday, 26-May-2007