Modeling Binary Fingerprint Descriptors With the Superposing Significant Interaction Rules (SSIR) Method

Modeling Binary Fingerprint Descriptors With the Superposing Significant Interaction Rules (SSIR) Method

Emili Besalú
DOI: 10.4018/IJQSPR.20200701.oa2
Article PDF Download
Open access articles are freely available for download


Recently, the superposing significant interaction rules (SSIR) method has been applied in several fields of QSPR to model and establish molecular rankings that correlate dichotomous properties. The origin of the method is in the field of combinatorial chemistry, but it has been shown that the procedure is fast, versatile, and that it can be applied in many other fields. In particular, an example is phospholipidosis modeling taking, as primary descriptors, the binary fingerprints of the molecules. This is the first time SSIR is used to treat this kind of descriptors. The performance achieved is similar to other results found in the literature and, in particular, to the results obtained by authors who considered the same molecular set and descriptors. One of the main advantages of SSIR is that the method acts as an automated variable selector. This allows it to be used almost immediately without prior selection of variables.
Article Preview

2. Material And Methods

The SSIR method is a variable selector based on the hypergeometric experiment (Mendenhall & Sincich, 1995). Briefly, given an urn containing red and green marbles, this probabilistic experiment consists of randomly selecting some of those marbles and, subsequently, assessing the probability associated with the distribution of red and green marbles which has been picked up. Specifically, the urn originally contains a marbles (b of them are green and the rest are red) and c are randomly selected. After the selection, one realizes that d of the marbles extracted are green. The probability for the described event follows the hypergeometric probability distribution:IJQSPR.20200701.oa2.m01 with dca and dba(1) where the minimum allowed value for d is max(0,c+b-a), and the maximum is min(b,c).

Complete Article List

Search this Journal:
Volume 9: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 8: 1 Issue (2023)
Volume 7: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 6: 4 Issues (2021)
Volume 5: 4 Issues (2020)
Volume 4: 4 Issues (2019)
Volume 3: 2 Issues (2018)
Volume 2: 2 Issues (2017)
Volume 1: 2 Issues (2016)
View Complete Journal Contents Listing