Dataset Information

Modeling temporal relationships in large scale clinical associations.

ABSTRACT:

Objective

We describe an approach for modeling temporal relationships in a large scale association analysis of electronic health record data. The addition of temporal information can inform hypothesis generation and help to explain the relationships. We applied this approach on a dataset containing 41.2 million time-stamped International Classification of Diseases, Ninth Revision (ICD-9) codes from 1.6 million patients.

Methods

We performed two independent analyses including a pairwise association analysis using a ?(2) test and a temporal analysis using a binomial test. Data were visualized using network diagrams and reviewed for clinical significance.

Results

We found nearly 400 000 highly associated pairs of ICD-9 codes with varying numbers of strong temporal associations ranging from ?1 day to ?10 years apart. Most of the findings were not considered clinically novel, although some, such as an association between Helicobacter pylori infection and diabetes, have recently been reported in the literature. The temporal analysis in our large cohort, however, revealed that diabetes usually preceded the diagnoses of H pylori, raising questions about possible cause and effect.

Discussion

Such analyses have significant limitations, some of which are due to known problems with ICD-9 codes and others to potentially incomplete data even at a health system level. Nevertheless, large scale association analyses with temporal modeling can help provide a mechanism for novel discovery in support of hypothesis generation.

Conclusions

Temporal relationships can provide an additional layer of meaning in identifying and interpreting clinical associations.

SUBMITTER: Hanauer DA

PROVIDER: S-EPMC3638191 | biostudies-literature | 2013 Mar-Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Modeling temporal relationships in large scale clinical associations.

Hanauer David A DA Ramakrishnan Naren N

Journal of the American Medical Informatics Association : JAMIA 20120927 2

<h4>Objective</h4>We describe an approach for modeling temporal relationships in a large scale association analysis of electronic health record data. The addition of temporal information can inform hypothesis generation and help to explain the relationships. We applied this approach on a dataset containing 41.2 million time-stamped International Classification of Diseases, Ninth Revision (ICD-9) codes from 1.6 million patients.<h4>Methods</h4>We performed two independent analyses including a pai ...[more]

PMID: 23019240

Similar Datasets

Project description:In response to stimuli, biological processes are tightly controlled by dynamic cellular signaling mechanisms. Reversible protein phosphorylation occurs on rapid time-scales (milliseconds to seconds), making it an ideal carrier of these signals. Advances in mass spectrometry-based proteomics have led to the identification of many tens of thousands of phosphorylation sites, yet for the majority of these the kinase is unknown and the underlying network topology of signaling networks therefore remains obscured. Identifying kinase substrate relationships (KSRs) is therefore an important goal in cell signaling research. Existing consensus sequence motif based prediction algorithms do not consider the biological context of KSRs, and are therefore insensitive to many other mechanisms guiding kinase-substrate recognition in cellular contexts. Here, we use temporal information to identify biologically relevant KSRs from Large-scale In Vivo Experiments (KSR-LIVE) in a data-dependent and automated fashion. First, we used available phosphorylation databases to construct a repository of existing experimentally-predicted KSRs. For each kinase in this database, we used time-resolved phosphoproteomics data to examine how its substrates changed in phosphorylation over time. Although substrates for a particular kinase clustered together, they often exhibited a different temporal pattern to the phosphorylation of the kinase. Therefore, although phosphorylation regulates kinase activity, our findings imply that substrate phosphorylation likely serve as a better proxy for kinase activity than kinase phosphorylation. KSR-LIVE can thereby infer which kinases are regulated within a biological context. Moreover, KSR-LIVE can also be used to automatically generate positive training sets for the subsequent prediction of novel KSRs using machine learning approaches. We demonstrate that this approach can distinguish between Akt and Rps6kb1, two kinases that share the same linear consensus motif, and provide evidence suggesting IRS-1 S265 as a novel Akt site. KSR-LIVE is an open-access algorithm that allows users to dissect phosphorylation signaling within a specific biological context, with the potential to be included in the standard analysis workflow for studying temporal high-throughput signal transduction data.

Dataset Information

Modeling temporal relationships in large scale clinical associations.

Objective

Methods

Results

Discussion

Conclusions

Publications

Modeling temporal relationships in large scale clinical associations.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets