SPStephen R. Piccolo

Stephen R. Piccolo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Stephen R. Piccolo has a total of 292 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Impacts of batch effects on the performance of machine learning classifiers across multiple studies

    by Preston Raab, William Evan Johnson, Stephen R Piccolo

    Precision medicine relies on accurate and generalizable predictions for patients across the spectrum of human diversity. Because capturing biological heterogeneity requires large sample sizes, researchers must often aggregate data from several experimental batches or independent studies. This integration allows for greater statistical power and diversity than a single study could provide, while avoiding the costs of generating massive new -omics datasets. Predictive models trained on these aggregated data are theoretically better equipped to detect subtle patterns that generalize to new data. However, this potential is frequently undermined by "batch effects"--systematic technical artifacts that can bias model training to predict experimental batches and shadow meaningful biological conditions. Models trained on data with batch effects can exhibit substantially degraded performance when applied to data from new batches. Statistical adjustment methods can mitigate these artifacts while preserving biological signals. To ensure these adjustments actually facilitate generalization, we emphasize the use of external, independent cohorts for rigorous validation. This chapter examines how batch effects impact predictions and compares various adjustment methods.

  • Open Access

    Using semantic search to find publicly available gene-expression datasets

    by Grace S Brown, James Wengler, Aaron Joyce S Fabelico, Abigail Muir, Anna Tubbs, Amanda Warren, Alexandra N Millett, Xinrui Xiang Yu, Paul Pavlidis, Sanja Rogic, et al.

    Motivation: Millions of high-throughput, molecular datasets have been shared in public repositories. Researchers can reuse such data to validate their own findings and explore novel questions. A frequent goal is to find multiple datasets that address similar research topics and to either combine them directly or integrate inferences from them. However, a major challenge is finding relevant datasets due to the vast number of candidates, inconsistencies in their descriptions, and a lack of semantic annotations. This challenge is first among the FAIR principles for scientific data. Here we focus on dataset discovery within Gene Expression Omnibus (GEO), a repository containing 100 000 s of data series. GEO supports queries based on keywords, ontology terms, and other annotations. However, reviewing these results is time-consuming and tedious, and it often misses relevant datasets. Results: We hypothesized that language models could address this problem by summarizing dataset descriptions as numeric representations (embeddings). Assuming a researcher has previously found some relevant datasets, we evaluated the potential to find additional relevant datasets. For six human medical conditions, we used 30 models to generate embeddings for datasets that human curators had previously associated with the conditions and identified other datasets with the most similar descriptions. This approach was often, but not always, more effective than GEO’s search engine. The top-performing models were trained on general corpora, used contrastive-learning strategies, and used relatively large embeddings. Our findings suggest that language models have the potential to improve dataset discovery, likely in combination with existing search tools. Availability and implementation: Our analysis code and a Web-based tool that enables others to use our methodology are available from https://github.com/srp33/GEO_NLP and https://github.com/srp33/GEOfinder3.0, respectively.


  • Open Access

    Using semantic search to find publicly available gene-expression datasets

    by Grace S. Brown, James Wengler, Aaron Joyce S. Fabelico, Abigail Muir, Anna Tubbs, Amanda Warren, Alexandra N. Millett, Xinrui Xiang Yu, Paul Pavlidis, Sanja Rogic, et al.

    Millions of high-throughput, molecular datasets have been shared in public repositories. have been shared in public repositories. Researchers can reuse such data to validate their own findings and explore novel questions. A frequent goal is to find multiple datasets that address similar research topics and to either combine them directly or integrate inferences from them. However, a major challenge is finding relevant datasets due to the vast number of candidates, inconsistencies in their descriptions, and a lack of semantic annotations. This challenge is first among the FAIR principles for scientific data. Here we focus on dataset discovery within Gene Expression Omnibus (GEO), a repository containing 100,000s of data series. GEO supports queries based on keywords, ontology terms, and other annotations. However, reviewing these results is time-consuming and tedious, and it often misses relevant datasets. We hypothesized that language models could address this problem by summarizing dataset descriptions as numeric representations (embeddings). Assuming a researcher has previously found some relevant datasets, we evaluated the potential to find additional relevant datasets. For six human medical conditions, we used 30 models to generate embeddings for datasets that human curators had previously associated with the conditions and identified other datasets with the most similar descriptions. This approach was often, but not always, more effective than GEO’s search engine. Our top-performing models were trained on general corpora, used contrastive-learning strategies, and used relatively large embeddings. Our findings suggest that language models have the potential to improve dataset discovery, perhaps in combination with existing search tools.

  • Open Access

    Translating short-form Python exercises to other programming languages using diverse prompting strategies

    by Stephen R Piccolo, Harlan P Stevens

    Background: With the increasing complexity and quantity of experimental and observational data, life scientists rely on programming to automate analyses, enhance reproducibility, and facilitate collaboration. Scripting languages like Python are often favored for their simplicity and flexibility, enabling researchers to focus primarily on high-level tasks. Compiled languages such as C++ and Rust offer greater efficiency, making them preferable for intensive or repeated computations. In educational settings, instructors may wish to teach both types of languages and thus translate content from one programming language to another. In research contexts, researchers may wish to implement their ideas in one language before translating the code to another. However, translating between programming languages requires significant effort, prompting our interest in using large language models (LLMs) for semi-automated code translation. Results: This study explores the use of an LLM (GPT-4) to translate 559 short-form programming exercises from Python into C++, Rust, Julia, and JavaScript. We used 3 prompting strategies—instructions only, code only, or both combined—and compared the translated code’s output against the Python code’s output. Translation success differed considerably by prompting strategy, and at least one of the strategies tested was effective for nearly every exercise. The highest overall success rate occurred for Rust (99.5%), followed by JavaScript (98.9%), C++ (97.9%), and Julia (95.0%). Conclusions: Our findings demonstrate that LLMs can effectively translate small-scale programming exercises between languages, reducing the need for manual rewriting. To support education and research, we have manually translated all exercises that were not translated successfully through automation, and we have made our translations freely available.


  • Open Access

    Opportunities and considerations for using artificial intelligence in bioinformatics education

    by Stephen R Piccolo, Aparna Nathan, Michelle D Brazas, Manoj Kandpal, Aida T Miró-Herrans, Adam J Kleinschmit, Susan McClatchy, Pertunia Mutheiwana, Dusanka Nikolic, Luciana I Gallo, et al.

    Artificial intelligence (AI) tools and techniques are undoubtedly being used in bioinformatics education, reflecting broader trends in education. However, many instructors and learners may be unaware of the full scope of potential uses for these tools within bioinformatics education, as well as effective practices for using them. Building on discussions held at the 6th Global Bioinformatics Education Summit, this perspective article provides insights about ways that AI might be used to generate or adapt instructional content, provide personalized help for learners, and automate assessment and grading. Additionally, we highlight AI skills that are important for bioinformatics learners to develop in order to effectively use AI as a bioinformatics learning tool. We highlight currently available tools in the quickly evolving AI landscape and suggest ways that instructors or learners might use such tools. Furthermore, we discuss key considerations and challenges associated with integrating AI into bioinformatics education, including ethical implications, potential biases, and the need to critically evaluate AI-generated content. Finally, we highlight the need for further research to better understand how AI tools are being used in practice and empower their effective and responsible use in bioinformatics education.


  • Open Access

    Comparison of Predictive Factors of Flu Vaccine Uptake Pre- and Post-COVID-19 Using the NIS-Teen Survey

    by Ty J. Skyles, Harlan P. Stevens, Spencer C. Davis, Acelan M. Obray, Dashiell S. Miner, Matthew J. East, Tyler Davis, Haley Hoelzer, Stephen R. Piccolo, Jamie L. Jensen, et al.

    Background: Seasonal influenza vaccination rates are very low among teenagers. Objectives: We used publicly available data from the NIS-Teen annual national immunization survey to explore factors that influence the likelihood of a teen receiving their seasonal flu shot. Methods: Traditional stepwise multivariable regression was used in tandem with machine learning to determine the predictive factors in teen vaccine uptake. Results and Conclusions: Age was the largest predictor, with older teens being much less likely to be vaccinated than younger teens (97.48% compared to 41.71%, p < 0.0001). Provider participation in government programs such as Vaccines for Children and the state vaccine registry positively impacts vaccine uptake (p < 0.0001). Identifying as non-Hispanic Black was a small, negative predictor of teen vaccine uptake (78.18% unvaccinated compared to 73.78% of White teens, p < 0.0001). The state quartile for COVID-19 vaccine uptake also strongly predicted flu vaccine uptake, with the upper quartile of state COVID-19 vaccine uptake being significantly more likely to also get vaccinated for influenza (76.96%, 74.94%, 74.55%, and 72.97%, p < 0.0001). Other significant factors are the number of providers, education of the mother, poverty status, and having a mixed provider facility type. Additionally, the multivariable regression analysis revealed little difference in the predictive factors of vaccine uptake between pre- and post-pandemic datasets.


  • Open Access

    Comparison of B-Cell Lupus and Lymphoma Using a Novel Immune Imbalance Transcriptomics Algorithm Reveals Potential Therapeutic Targets

    by Naomi Rapier-Sharman, Sehi Kim, Madelyn Mudrow, Michael T. Told, Lane Fischer, Liesl Fawson, Joseph Parry, Brian D. Poole, Kim L. O’neill, Stephen R. Piccolo, et al.

    Background/Objectives: Systemic lupus erythematosus (lupus) and B-cell lymphoma (lymphoma) co-occur at higher-than-expected rates and primarily depend on B cells for their pathology. These observations implicate shared inflammation-related B cell molecular mechanisms as a potential cause of co-occurrence. Methods: We consequently implemented a novel Immune Imbalance Transcriptomics (IIT) algorithm and applied IIT to lupus, lymphoma, and healthy B cell RNA-sequencing (RNA-seq) data to find shared and contrasting mechanisms that are potential therapeutic targets. Results: We observed 7143 significantly dysregulated genes in both lupus and lymphoma. Of those genes, we found 5137 to have a significant immune imbalance, defined as a significant dysregulation by both diseases, as analyzed by IIT. Gene Ontology (GO) term and pathway enrichment of the IIT genes yielded immune-related “Neutrophil Degranulation” and “Adaptive Immune System”, which validates that the IIT algorithm isolates biologically relevant genes in immunity and inflammation. We found that 344 IIT gene products are known targets for established and/or repurposed drugs. Among our results, we found 48 known and 296 novel lupus targets, along with 151 known and 193 novel lymphoma targets. Known disease drug targets in our IIT results further validate that IIT isolates genes with disease-relevant mechanisms. Conclusions: We anticipate the IIT algorithm, together with the shared and contrasting gene mechanisms uncovered here, will contribute to the development of immune-related therapeutic options for lupus and lymphoma patients.


  • Open Access

    Identifying images in the biology literature that are problematic for people with a color-vision deficiency

    by Harlan P Stevens, Carly V Winegar, Arwen F Oakley, Stephen R Piccolo

    To help maximize the impact of scientific journal articles, authors must ensure that article figures are accessible to people with color-vision deficiencies (CVDs), which affect up to 8% of males and 0.5% of females. We evaluated images published in biology- and medicine-oriented research articles between 2012 and 2022. Most included at least one color contrast that could be problematic for people with deuteranopia (‘deuteranopes’), the most common form of CVD. However, spatial distances and within-image labels frequently mitigated potential problems. Initially, we reviewed 4964 images from eLife, comparing each against a simulated version that approximated how it might appear to deuteranopes. We identified 636 (12.8%) images that we determined would be difficult for deuteranopes to interpret. Our findings suggest that the frequency of this problem has decreased over time and that articles from cell-oriented disciplines were most often problematic. We used machine learning to automate the identification of problematic images. For a hold-out test set from eLife (n=879), a convolutional neural network classified the images with an area under the precision-recall curve of 0.75. The same network classified images from PubMed Central (n=1191) with an area under the precision-recall curve of 0.39. We created a Web application (https://bioapps.byu.edu/colorblind_image_tester); users can upload images, view simulated versions, and obtain predictions. Our findings shed new light on the frequency and nature of scientific images that may be problematic for deuteranopes and motivate additional efforts to increase accessibility.


  • Open Access

    A Comprehensive Meta-Analysis of Breast Cancer Gene Expression

    by Ifeanyichukwu O. Nwosu, Stephen R. Piccolo

    Background: Triple-negative breast cancers (TNBC) occur more frequently in African Americans and are associated with worse outcomes when compared to other subtypes of breast cancer. These cancers lack expression of estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) and have limited treatment options. To shed light on mechanisms behind these differences and suggest novel treatments, we used a meta-analytic approach to identify gene expression differences in breast tumors for people with self-reported African or European ancestry; additionally, we compared gene expression levels based on ER, PR, HER2 and TNBC status. Methods: After gathering and standardizing gene expression data and metadata from 106 datasets (representing 27,000 samples), we identified genes that were expressed differently between these groups via random-effects meta-analyses. To evaluate the robustness of these gene lists, we devised a novel computational methodology that uses cross validation and classification. We also computed overlaps between the most significant genes and known signaling pathways. Results: Using a false discovery rate threshold of 0.05, we identified genes that are known to play a significant role in their respective breast cancer subtypes (e.g., ESR1 for ER status and ERBB2 for HER2 status), thus confirming the validity of our findings. We also discovered genes that have not been reported previously and may be new targets for breast cancer therapy. GATA3 , CA12 , TBC1D9 , XBP1 and FOXA1 were among the most significant genes for ER, PR, and TNBC. However, none of these genes overlapped with HER2 status, supporting prior research that HER2 tumors are mechanistically different from endocrine breast cancers. The genes identified from the race meta-analysis—including DNAJC15 , HLA-DPA1 , STAP2 , CEP68 , MOGS —have not been associated previously with race-specific breast-cancer outcomes, highlighting a potential area of further research. Conclusions: We have carried out a large meta-analysis of breast cancer gene expression data, identifying novel genes that may serve as potential biomarkers for breast cancer in diverse populations. We have also developed a computational method that identifies gene sets small enough to be analyzed and explored in future studies. This method has the potential to be applied to other cancers.


  • Open Access

    Identifying images in the biology literature that are problematic for people with a color-vision deficiency

    by Harlan P Stevens, Carly V Winegar, Arwen F Oakley, Stephen R Piccolo

    To help maximize the impact of scientific journal articles, authors must ensure that article figures are accessible to people with color-vision deficiencies (CVDs), which affect up to 8% of males and 0.5% of females. We evaluated images published in biology-and medicine-oriented research articles between 2012 and 2022. Most included at least one color contrast that could be problematic for people with deuteranopia (“deuteranopes”), the most common form of CVD. However, spatial distances and within-image labels frequently mitigated potential problems. Initially, we reviewed 4,964 images from eLife, comparing each against a simulated version that approximated how it might appear to deuteranopes. We identified 636 (12.8%) images that we determined would be difficult for deuteranopes to interpret. Our findings suggest that the frequency of this problem has decreased over time and that articles from cell-oriented disciplines were most often problematic. We used machine learning to automate the identification of problematic images. For hold-out test sets from eLife (n = 879) and PubMed Central (n = 1,191), a convolutional neural network classified the images with areas under the precision-recall curve of 0.75 and 0.38, respectively. We created a Web application (https://bioapps.byu.edu/colorblind_image_tester); users can upload images, view simulated versions, and obtain predictions. Our findings shed new light on the frequency and nature of scientific images that may be problematic for deuteranopes and motivate additional efforts to increase accessibility.