kenmcgarry Goto Github PK

followers: 9.0 following: 5.0 repos: 24.0 gists: 0.0

Name: Ken McGarry

Type: User

Company: University of Sunderland

Bio: One man's effort in creating knowledge from data.

Location: UK, England

Blog: http://fas-web.sunderland.ac.uk/~cs0kmc/

Ken McGarry's Projects

atc

Anatomical Therapeutic Codes (ATC) are a drug classification system which is extensively used in the field of drug development research.

Bayesian surprise is the result of mismatches between our expectations and actual results, hence the degree of surprise or anomalousness attached to a pattern will vary with respect to these differences. The implication of obtaining large surprise values identifies those patterns likely to be useful and interesting to the user.

biologicalplausibility

Bioinformatics algorithms need the ability to assess the relevance and biological plausibility of their discoveries.

communitydetection

Community network discovery by weighting a random walk algorithm with ontological information

complexnetworks

complex network theory for identifying drug-target proteins. Undergoing some revisions and bug fixes.

disease-modules

The new science of complex networks has revealed the dynamic nature of diseases through shared genes and mechanisms.

drugsideeffects

Drugs with similar side-effects are potential candidates for use elsewhere, the supposition is that similar side-effects may be caused by drugs targeting similar proteins.

edgetics

Complex networks of SNP's - work in progress

go-slim

comparision of GO-slim ontologies for building classifiers

heuristics

Heuristics are often described as rules of thumb or short cuts to useful solutions that avoid lengthy or complex calculations.

hypothesis

Hypothesis creation and testing in a data mining domain. We develop a reasoning system that is seeded with a base level of data mining knowledge and is capable of expanding, modifying and updating this knowledge with new experiences.

interestingpatterns

One of the most insightful definitions of data mining states that to be truly successful data mining should be “the nontrivial process of identifying valid, novel, potentially useful, and ultimately comprehensible knowledge from databases”

motif

Using REGEX to seek patterns (motifs)

neuralnetworks

Using SVM, MLP and RBF for predicting protein targets

paradox

Paradox detection in data for knowledge discovery

qsar-hiv

Modelling QSAR compound data for affinity to binding with GP120/CD4 proteins.

reinforcement

Integrating reinforcement learning within a cognitive framework for pattern detection

sbm

In this work we integrate complex networks and stochastic block models with heuristic reasoning for the purposes of data mining interesting patterns.

textminer

Turning messy data into tidy data. The majority of human knowledge and experience is in the form of the written word (messy data) and not structured databases (tidy data) which is required for machine learning algorithms.

ukci-2014-diseasenetworks

The R work described in our conference paper presented at UKCI-2014, in Bradford, 8th-10th Sept

ukci2015-side-effects

The R work described in our conference paper presented at UKCI-2015 in Exeter, 7th-9th Sept. Drug development is a lengthy and highly costly endeavor, often with limited success and high risk. The objective of drug repositioning is to apply existing drugs to different diseases or medical conditions than the original target, and thus alleviate to a certain extent the time and cost expended.

ukci2016-edgetics

The R work described in conference paper #1 presented at UKCI-2016 in Lancaster 7th-9th Sept. Complex networks are a graph theoretic method that can model genetic mutations, in particular single nucleotide polymorphisms (snp’s) which are genetic variations that only occur at single position in a DNA sequence.

ukci2016-mcl

The R work described in conference paper #2 presented at UKCI-2016 in Lancaster 7th-9th Sept. The detection of protein complexes is an important research problem in bioinformatics, which may help increase our understanding of the biological functions of proteins inside our body.

ukci2017-ar

Our knowledge of drug-to-drug interactions, side-effects and disease comorbidity is derived from healthcare record systems and these are now starting to receive increased attention as a way of improving public health and drug safety.

kenmcgarry Goto Github PK

Ken McGarry's Projects

Recommend Projects

Recommend Topics

Recommend Org