Dataset Information

Optimizing artificial neural network models for metabolomics and systems biology: an example using HPLC retention index data.

ABSTRACT: Artificial Neural Networks (ANN) are extensively used to model 'omics' data. Different modeling methodologies and combinations of adjustable parameters influence model performance and complicate model optimization.We evaluated optimization of four ANN modeling parameters (learning rate annealing, stopping criteria, data split method, network architecture) using retention index (RI) data for 390 compounds. Models were assessed by independent validation (I-Val) using newly measured RI values for 1492 compounds.The best model demonstrated an I-Val standard error of 55 RI units and was built using a Ward's clustering data split and a minimally nonlinear network architecture. Use of validation statistics for stopping and final model selection resulted in better independent validation performance than the use of test set statistics.

SUBMITTER: Hall LM

PROVIDER: S-EPMC4462168 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Optimizing artificial neural network models for metabolomics and systems biology: an example using HPLC retention index data.

Hall L Mark LM Hill Dennis W DW Menikarachchi Lochana C LC Chen Ming-Hui MH Hall Lowell H LH Grant David F DF

Bioanalysis 20150101 8

<h4>Background</h4>Artificial Neural Networks (ANN) are extensively used to model 'omics' data. Different modeling methodologies and combinations of adjustable parameters influence model performance and complicate model optimization.<h4>Methodology</h4>We evaluated optimization of four ANN modeling parameters (learning rate annealing, stopping criteria, data split method, network architecture) using retention index (RI) data for 390 compounds. Models were assessed by independent validation (I-Va ...[more]

PMID: 25966007

Similar Datasets

Project description:Liquid chromatography coupled with electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) is a major analytical technique used for nontargeted identification of metabolites in biological fluids. Typically, in LC-ESI-MS/MS based database assisted structure elucidation pipelines, the exact mass of an unknown compound is used to mine a chemical structure database to acquire an initial set of possible candidates. Subsequent matching of the collision induced dissociation (CID) spectrum of the unknown to the CID spectra of candidate structures facilitates identification. However, this approach often fails because of the large numbers of potential candidates (i.e., false positives) for which CID spectra are not available. To overcome this problem, CID fragmentation predication programs have been developed, but these also have limited success if large numbers of isomers with similar CID spectra are present in the candidate set. In this study, we investigated the use of a retention index (RI) predictive model as an orthogonal method to help improve identification rates. The model was used to eliminate candidate structures whose predicted RI values differed significantly from the experimentally determined RI value of the unknown compound. We tested this approach using a set of ninety-one endogenous metabolites and four in silico CID fragmentation algorithms: CFM-ID, CSI:FingerID, Mass Frontier, and MetFrag. Candidate sets obtained from PubChem and the Human Metabolite Database (HMDB) were ranked with and without RI filtering followed by in silico spectral matching. Upon RI filtering, 12 of the ninety-one metabolites were eliminated from their respective candidate sets, i.e., were scored incorrectly as negatives. For the remaining seventy-nine compounds, we show that RI filtering eliminated an average of 58% from PubChem candidate sets. This resulted in an approximately 2-fold improvement in average rankings when using CFM-ID, Mass Frontier, and MetFrag. In addition, RI filtering slightly increased the occurrence of number one rankings for all 4 fragmentation algorithms. However, RI filtering did not significantly improve average rankings when HMDB was used as the candidate database, nor did it significantly improve average rankings when using CSI:FingerID. Overall, we show that the current RI model incorrectly eliminated more true positives (12) than were expected (4-5) on the basis of the filtering method. However, it slightly improved the number of correct first place rankings and improved overall average rankings when using CFM-ID, Mass Frontier, and MetFrag.

Project description:The goal of many metabolomic studies is to identify the molecular structure of endogenous molecules that are differentially expressed among sampled or treatment groups. The identified compounds can then be used to gain an understanding of disease mechanisms. Unfortunately, despite recent advances in a variety of analytical techniques, small molecule (<1000 Da) identification remains difficult. Rarely can a chemical structure be determined from experimental "features" such as retention time, exact mass, and collision induced dissociation spectra. Thus, without knowing structure, biological significance remains obscure. In this study, we explore an identification method in which the measured exact mass of an unknown is used to query available chemical databases to compile a list of candidate compounds. Predictions are made for the candidates using models of experimental features that have been measured for the unknown. The predicted values are used to filter the candidate list by eliminating compounds with predicted values substantially different from the unknown. The intent is to reduce the list of candidates to a reasonable number that can be obtained and measured for confirmation. To facilitate this exploration, we measured data and created models for two experimental features; MS Ecom₅₀ (the energy in electronvolts required to fragment 50% of a selected precursor ion) and HPLC retention index. Using a data set of 52 compounds, Ecom₅₀ models were developed based on both Molconn and CODESSA structural descriptors. These models gave r² values of 0.89 to 0.94 depending on the number of inputs, the modeling algorithm chosen, and whether neutral or protonated structures were used. The retention index model was developed with 400 compounds using a back-propagation artificial neural network and 33 Molconn structure descriptors. External validation gave a v² = 0.87 and standard error of 38 retention index units. As a test of the validity of the filtering approach, the Ecom₅₀ and retention index models, along with exact mass and collision induced dissociation spectra matching, were used to identify 1,3-dicyclohexylurea in human plasma. This compound was not previously known to exist in human biofluids and its elemental formula was identical to 315 other candidate compounds downloaded from PubChem. These results suggest that the use of Ecom₅₀ and retention index predictive models can improve nontargeted metabolite structure identification using HPLC/MS derived structural features.

Project description:INTRODUCTION:The unexpected identification of a neural tube defect (NTD) safety signal with preconception dolutegravir (DTG) exposure in the Botswana Tsepamo birth outcomes study brought into sharp focus the need for reliable data on use of new antiretrovirals in pregnancy, improved pharmacovigilance systems to evaluate safety of new drugs being introduced into populations including women of reproductive potential, and balanced risk-benefit messaging when a safety signal is identified. DISCUSSION:The Tsepamo study NTD safety signal and accompanying regulatory responses led to uncertainty about the most appropriate approach to DTG use among women of reproductive potential, affecting global DTG roll-out plans, and limiting DTG use in adolescent girls and women. It also revealed a tension between a public health approach to antiretroviral treatment (ART) and individual choice, and highlighted difficulties interpreting and messaging an unexpected safety signal with uncertainty about risk. This difficulty was compounded by the lack of high-quality data on pregnancy outcomes from women receiving ART outside the Tsepamo surveillance sites and countries other than Botswana, resulting in a prolonged period of uncertainty while data on additional exposures are evaluated to refute or confirm the initial safety signal. We discuss principles for evaluating and introducing new drugs in the general population that would ensure collection of appropriate data to inform drug safety in adolescent girls and women of reproductive potential and minimize confusion about drug use in this population when a safety signal is identified. CONCLUSIONS:The response to a signal suggesting a possible safety risk for a drug used in pregnancy or among women who may become pregnant needs to be rapid and comprehensive. It requires the existence of appropriately designed surveillance systems with broad population coverage; data analyses that examine risk-benefit trade-offs in a variety of contexts; guidance to transform this risk-benefit balance into effective and agreed-upon policy; involvement of the affected community and other key stakeholders; and a communication plan for all levels of knowledge and complexity. Implementation of this proposed framework for responding to safety signals is needed to ensure that any drug used in pregnancy can be rapidly and appropriately evaluated should a serious safety alert arise.

Dataset Information

Optimizing artificial neural network models for metabolomics and systems biology: an example using HPLC retention index data.

Publications

Optimizing artificial neural network models for metabolomics and systems biology: an example using HPLC retention index data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets