Ping Gong

Dr. Ping Gong

orcid

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 December 2017 in BMC Bioinformatics
Verified
Journal Article
Open access
PDF file
AMAndrew Maxwell
Andrew Maxwell
RLRun-Zhi Li
Run-Zhi Li
BYBei Yang
Bei Yang
HWHeng Weng
Heng Weng
AOAihua Ou
Aihua Ou
CZChaoyang (Joe) Zhang
Chaoyang (Joe) Zhang
Multi-label classification of data remains to be a challenging problem. Because of the complexity of the data, it is sometimes difficult to infer information about classes that are not mutually exclusive. For medical data, patients could have symptoms of multiple different diseases at the same time and it is important to develop tools that help to identify problems early. Intelligent health risk prediction models built with deep learning architectures offer a powerful tool for physicians to identify patterns in patient data that indicate risks associated with certain types of chronic diseases. Physical examination records of 110,300 anonymous patients were used to predict diabetes, hypertension, fatty liver, a combination of these three chronic diseases, and the absence of disease (8 classes in total). The dataset was split into training (90%) and testing (10%) sub-datasets. Ten-fold cross validation was used to evaluate prediction accuracy with metrics such as precision, recall, and F-score. Deep Learning (DL) architectures were compared with standard and state-of-the-art multi-label classification methods. Preliminary results suggest that Deep Neural Networks (DNN), a DL architecture, when applied to multi-label classification of chronic diseases, produced accuracy that was comparable to that of common methods such as Support Vector Machines. We have implemented DNNs to handle both problem transformation and algorithm adaption type multi-label methods and compare both to see which is preferable. Deep Learning architectures have the potential of inferring more information about the patterns of physical examination data than common classification methods. The advanced techniques of Deep Learning can be used to identify the significance of different features from physical examination data as well as to learn the contributions of each feature that impact a patient’s risk for chronic diseases. However, accurate prediction of chronic disease risks remains a challenging problem that warrants further studies.

27 October 2020 in Journal of Cheminformatics
Verified
Journal Article
Open access
PDF file
GIGabriel Idakwo
Gabriel Idakwo
STSundar Thangapandian
Sundar Thangapandian
JLJoseph Luttrell
Joseph Luttrell
YLYan Li
Yan Li
NWNan Wang
Nan Wang
PGPing Gong
Ping Gong
The specificity of toxicant-target biomolecule interactions lends to the very imbalanced nature of many toxicity datasets, causing poor performance in Structure–Activity Relationship (SAR)-based chemical classification. Undersampling and oversampling are representative techniques for handling such an imbalance challenge. However, removing inactive chemical compound instances from the majority class using an undersampling technique can result in information loss, whereas increasing active toxicant instances in the minority class by interpolation tends to introduce artificial minority instances that often cross into the majority class space, giving rise to class overlapping and a higher false prediction rate. In this study, in order to improve the prediction accuracy of imbalanced learning, we employed SMOTEENN, a combination of Synthetic Minority Over-sampling Technique (SMOTE) and Edited Nearest Neighbor (ENN) algorithms, to oversample the minority class by creating synthetic samples, followed by cleaning the mislabeled instances. We chose the highly imbalanced Tox21 dataset, which consisted of 12 in vitro bioassays for > 10,000 chemicals that were distributed unevenly between binary classes. With Random Forest (RF) as the base classifier and bagging as the ensemble strategy, we applied four hybrid learning methods, i.e., RF without imbalance handling (RF), RF with Random Undersampling (RUS), RF with SMOTE (SMO), and RF with SMOTEENN (SMN). The performance of the four learning methods was compared using nine evaluation metrics, among which F1 score, Matthews correlation coefficient and Brier score provided a more consistent assessment of the overall performance across the 12 datasets. The Friedman’s aligned ranks test and the subsequent Bergmann-Hommel post hoc test showed that SMN significantly outperformed the other three methods. We also found that a strong negative correlation existed between the prediction accuracy and the imbalance ratio (IR), which is defined as the number of inactive compounds divided by the number of active compounds. SMN became less effective when IR exceeded a certain threshold (e.g., > 28). The ability to separate the few active compounds from the vast amounts of inactive ones is of great importance in computational toxicology. This work demonstrates that the performance of SAR-based, imbalanced chemical toxicity classification can be significantly improved through the use of data rebalancing.

14 March 2019 in BMC Bioinformatics
Verified
Journal Article
Open access
PDF file
BPBohu Pan
Bohu Pan
RKRebecca Kusko
Rebecca Kusko
WXWenming Xiao
Wenming Xiao
YZYuanting Zheng
Yuanting Zheng
ZLZhichao Liu
Zhichao Liu
HHHuixiao Hong
Huixiao Hong
Reference genome selection is a prerequisite for successful analysis of next generation sequencing (NGS) data. Current practice employs one of the two most recent human reference genome versions: HG19 or HG38. To date, the impact of genome version on SNV identification has not been rigorously assessed. We conducted analysis comparing the SNVs identified based on HG19 vs HG38, leveraging whole genome sequencing (WGS) data from the genome-in-a-bottle (GIAB) project. First, SNVs were called using 26 different bioinformatics pipelines with either HG19 or HG38. Next, two tools were used to convert the called SNVs between HG19 and HG38. Lastly we calculated conversion rates, analyzed discordant rates between SNVs called with HG19 or HG38, and characterized the discordant SNVs. The conversion rates from HG38 to HG19 (average 95%) were lower than the conversion rates from HG19 to HG38 (average 99%). The conversion rates varied slightly among the various calling pipelines. Around 1.5% SNVs were discordantly converted between HG19 or HG38. The conversions from HG38 to HG19 had more SNVs which failed conversion and more discordant SNVs than the opposite conversion (HG19 to HG38). Most of the discordant SNVs had low read depth, were low confidence SNVs as defined by GIAB, and/or were predominated by G/C alleles (52% observed versus 42% expected). A significant number of SNVs could not be converted between HG19 and HG38. Based on careful review of our comparisons, we recommend HG38 (the newer version) for NGS SNV analysis. To summarize, our findings suggest caution when translating identified SNVs between different versions of the human reference genome.

13 August 2019 in Frontiers in Physiology
Verified
Journal Article
Open access
GIGabriel Idakwo
Gabriel Idakwo
STSundar Thangapandian
Sundar Thangapandian
JLJoseph Luttrell
Joseph Luttrell
ZZZhaoxian Zhou
Zhaoxian Zhou
CZChaoyang (Joe) Zhang
Chaoyang (Joe) Zhang
PGPing Gong
Ping Gong
Deep learning (DL) has attracted the attention of computational toxicologists as it offers a potentially greater power for in silico predictive toxicology than existing shallow learning algorithms. However, contradicting reports have been documented. To further explore the advantages of DL over shallow learning, we conducted this case study using two cell-based androgen receptor (AR) activity datasets with 10K chemicals generated from the Tox21 program. A nested double-loop cross-validation approach was adopted along with a stratified sampling strategy for partitioning chemicals of multiple AR activity classes (i.e., agonist, antagonist, inactive, and inconclusive) at the same distribution rates amongst the training, validation and test subsets. Deep neural networks (DNN) and random forest (RF), representing deep and shallow learning algorithms, respectively, were chosen to carry out structure-activity relationship-based chemical toxicity prediction. Results suggest that DNN significantly outperformed RF (p < 0.001, ANOVA) by 22–27% for four metrics (precision, recall, F-measure, and AUPRC) and by 11% for another (AUROC). Further in-depth analyses of chemical scaffolding shed insights on structural alerts for AR agonists/antagonists and inactive/inconclusive compounds, which may aid in future drug discovery and improvement of toxicity prediction modeling.

12 June 2021 in Insects
Verified
Journal Article
Open access
PDF file
Nathan HarmsNathan Harms
Nathan Harms
IKIan A. Knight
Ian A. Knight
PPPaul D. Pratt
Paul D. Pratt
ARAngelica Reddy
Angelica Reddy
AMAbhishek Mukherjee
Abhishek Mukherjee
RDRODRIGO DIAZ
RODRIGO DIAZ
Many weed biological control programs suffer from large-scale spatial variation in success due to restricted distributions or abundances of agents in temperate climates. For some of the world’s worst aquatic weeds, agents are established but overwintering conditions limit their survival in higher latitudes or elevations. The resulting need is for new or improved site- or region-specific biological control tools. Here, we review this challenge with a focus on low-temperature limitations of agents and propose a roadmap for improving success. Investigations across spatial scales, from global (e.g., foreign exploration), to local (selective breeding), to individual organisms (molecular modification), are discussed. A combination of traditional (foreign) and non-traditional (introduced range) exploration may lead to the discovery and development of better-adapted agent genotypes. A multivariate approach using ecologically relevant metrics to quantify and compare cold tolerance among agent populations is likely required. These data can be used to inform environmental niche modeling combined with mechanistic modeling of species’ fundamental climate niches and life histories to predict where, when, and at what abundance agents will occur. Finally, synthetic and systems biology approaches in conjunction with advanced modern genomics, gene silencing and gene editing technologies may be used to identify and alter the expression of genes enhancing cold tolerance, but this technology in the context of weed biological control has not been fully explored.

7 October 2023 in Biology
Verified
Journal Article
Open access
PDF file
LHLei Huang
Lei Huang
MSMeng Song
Meng Song
HSHui Shen
Hui Shen
HHHuixiao Hong
Huixiao Hong
PGPing Gong
Ping Gong
CZChaoyang (Joe) Zhang
Chaoyang (Joe) Zhang
One common problem in omics data analysis is missing values, which can arise due to various reasons, such as poor tissue quality and insufficient sample volumes. Instead of discarding missing values and related data, imputation approaches offer an alternative means of handling missing data. However, the imputation of missing omics data is a non-trivial task. Difficulties mainly come from high dimensionality, non-linear or non-monotonic relationships within features, technical variations introduced by sampling methods, sample heterogeneity, and the non-random missingness mechanism. Several advanced imputation methods, including deep learning-based methods, have been proposed to address these challenges. Due to its capability of modeling complex patterns and relationships in large and high-dimensional datasets, many researchers have adopted deep learning models to impute missing omics data. This review provides a comprehensive overview of the currently available deep learning-based methods for omics imputation from the perspective of deep generative model architectures such as autoencoder, variational autoencoder, generative adversarial networks, and Transformer, with an emphasis on multi-omics data imputation. In addition, this review also discusses the opportunities that deep learning brings and the challenges that it might face in this field.