Research Identifiers
ORCID iD:
https://orcid.org/0000-0003-1696-1781
Keywords
Literature-based discovery; causal modeling; causal feature selection;
Funding (4)
Using the literature to build causal models of retrospective observational data ✓ NIH
2023-08 to 2026-07 | Award
United States National Library of Medicine (Bethesda, Maryland, US)
Homepage URL: https://reporter.nih.gov/search/agkD0LqiWk-OD1wThSJt3Q/project-details/10879451
GRANT_NUMBER: R00LM013367
Show more detailOrganization identifiers
United States National Library of Medicine: Bethesda, Maryland, US
📄 Project Abstract (from NIH)
Health data contain a wealth of information for research. Health data, such as found in electronic health
records (EHRs), allow for the identification links between health events, such as drug exposures and side-
effects. Some of these links indicate stable dependencies deemed as causes. Causal insight allows reverse-
engineering disease. If confounding is not addressed, it will be difficult to distinguish causative from correlative
links. Our approach is to identify confounders explicitly. Graphical causal modeling (GCMs) can discover
causal links from data and prior knowledge. GCMs summarize causal links between variables. Automated
selection of variables would allow GCMs to scale and yield more insight from data. Literature-based discovery
(LBD) methods were developed to identify links between concepts in the literature. Advanced methods permit
the search for concepts linked to each other through specific verbs, e.g., “causes”, “treats”. Our hypothesis is
that we can exploit structured knowledge extracted from the literature to inform GCMs. In prior work, we found
that LBD + GCM was better at identifying side-effects in EHR data than traditional methods. Compared to
methods which use solely data, we hypothesize that our method will increase the ability to detect causal
relationships from EHR data. The first aim is to determine the extent to which LBD-informed GCM improves the
identification of causal links for drug safety. We will build LBD-informed GCMs using publicly available
reference datasets for drug safety. These reference datasets contain drug/side-effect pairs for performance
benchmarking. (A) Test the ability of GCM algorithms to identify known causal links solely using data. We will
systematically evaluate GCM algorithms based on their ability to re-discover causal links in a reference
standard. Results will guide our studies on how GCM can be tuned. (B) Determine the effect of adding different
subsets of LBD-derived information to GCMs at identifying drug side-effects. We will build causal models using
increasing numbers confounders. The second aim is to test the ability of LBD built with disease-specific
literature to improve the relevance of LBD derived confounders for Alzheimer's Disease (AD). We chose AD for
its high prevalence and relative lack of effective pharmacologic treatment. (A) Compare LBD strategies in a
disease-specific setting. We will test LBD variants using disease-specific literature or with LBD lacking subject-
matter restrictions. (B) Define the ability of robust LBD-informed GCM to validate drug repurposing candidates
for treating AD symptoms. We will test the ability of advanced methods to iteratively resolve hidden latent
confounding, when detected, to improve effect estimates. The fulfillment of these aims will yield new methods
to combine insights from the literature with causal modeling to uncover causal relationships of drug exposures
on adverse events and on beneficial outcomes.
records (EHRs), allow for the identification links between health events, such as drug exposures and side-
effects. Some of these links indicate stable dependencies deemed as causes. Causal insight allows reverse-
engineering disease. If confounding is not addressed, it will be difficult to distinguish causative from correlative
links. Our approach is to identify confounders explicitly. Graphical causal modeling (GCMs) can discover
causal links from data and prior knowledge. GCMs summarize causal links between variables. Automated
selection of variables would allow GCMs to scale and yield more insight from data. Literature-based discovery
(LBD) methods were developed to identify links between concepts in the literature. Advanced methods permit
the search for concepts linked to each other through specific verbs, e.g., “causes”, “treats”. Our hypothesis is
that we can exploit structured knowledge extracted from the literature to inform GCMs. In prior work, we found
that LBD + GCM was better at identifying side-effects in EHR data than traditional methods. Compared to
methods which use solely data, we hypothesize that our method will increase the ability to detect causal
relationships from EHR data. The first aim is to determine the extent to which LBD-informed GCM improves the
identification of causal links for drug safety. We will build LBD-informed GCMs using publicly available
reference datasets for drug safety. These reference datasets contain drug/side-effect pairs for performance
benchmarking. (A) Test the ability of GCM algorithms to identify known causal links solely using data. We will
systematically evaluate GCM algorithms based on their ability to re-discover causal links in a reference
standard. Results will guide our studies on how GCM can be tuned. (B) Determine the effect of adding different
subsets of LBD-derived information to GCMs at identifying drug side-effects. We will build causal models using
increasing numbers confounders. The second aim is to test the ability of LBD built with disease-specific
literature to improve the relevance of LBD derived confounders for Alzheimer's Disease (AD). We chose AD for
its high prevalence and relative lack of effective pharmacologic treatment. (A) Compare LBD strategies in a
disease-specific setting. We will test LBD variants using disease-specific literature or with LBD lacking subject-
matter restrictions. (B) Define the ability of robust LBD-informed GCM to validate drug repurposing candidates
for treating AD symptoms. We will test the ability of advanced methods to iteratively resolve hidden latent
confounding, when detected, to improve effect estimates. The fulfillment of these aims will yield new methods
to combine insights from the literature with causal modeling to uncover causal relationships of drug exposures
on adverse events and on beneficial outcomes.
👤 Principal Investigator(s) (from NIH)
Scott Alexander Malec
🏛️ Recipient Organization (from NIH)
UNIVERSITY OF NEW MEXICO HEALTH SCIS CTR (ALBUQUERQUE, NM, UNITED STATES)
📅 Project Dates (from NIH)
Start: 2021-08-01T00:00:00
End: 2026-07-31T00:00:00
End: 2026-07-31T00:00:00
💰 Award Amount (from NIH)
$248,671
📊 Fiscal Year (from NIH)
2025
🏷️ Activity Code (from NIH)
R00
🔢 Project Number (from NIH)
5R00LM013367-05
🔗 Full Project Record (from NIH)
Added
2023-11-16
Last modified
2023-11-16
Source:
Scott Alexander Malec
| ✓ Enriched from NIH
Scott Alexander Malec
| ✓ Enriched from NIH
Using the literature to build causal models of retrospective observational data ✓ NIH
2021-08-01 to 2023-07-31 | Grant
United States National Library of Medicine (Bethesda, US)
Homepage URL: https://app.dimensions.ai/details/grant/grant.9844339
GRANT_NUMBER: K99LM013367
Show more detailOrganization identifiers
United States National Library of Medicine: Bethesda, US
Funding project translated title
Funding project translated title
(en)
Using the literature to build causal models of retrospective observational data
📄 Project Abstract (from NIH)
Health data contain a wealth of information for research. Health data, such as found in electronic health
records (EHRs), allow for the identification links between health events, such as drug exposures and side-
effects. Some of these links indicate stable dependencies deemed as causes. Causal insight allows reverse-
engineering disease. If confounding is not addressed, it will be difficult to distinguish causative from correlative
links. Our approach is to identify confounders explicitly. Graphical causal modeling (GCMs) can discover
causal links from data and prior knowledge. GCMs summarize causal links between variables. Automated
selection of variables would allow GCMs to scale and yield more insight from data. Literature-based discovery
(LBD) methods were developed to identify links between concepts in the literature. Advanced methods permit
the search for concepts linked to each other through specific verbs, e.g., “causes”, “treats”. Our hypothesis is
that we can exploit structured knowledge extracted from the literature to inform GCMs. In prior work, we found
that LBD + GCM was better at identifying side-effects in EHR data than traditional methods. Compared to
methods which use solely data, we hypothesize that our method will increase the ability to detect causal
relationships from EHR data. The first aim is to determine the extent to which LBD-informed GCM improves the
identification of causal links for drug safety. We will build LBD-informed GCMs using publicly available
reference datasets for drug safety. These reference datasets contain drug/side-effect pairs for performance
benchmarking. (A) Test the ability of GCM algorithms to identify known causal links solely using data. We will
systematically evaluate GCM algorithms based on their ability to re-discover causal links in a reference
standard. Results will guide our studies on how GCM can be tuned. (B) Determine the effect of adding different
subsets of LBD-derived information to GCMs at identifying drug side-effects. We will build causal models using
increasing numbers confounders. The second aim is to test the ability of LBD built with disease-specific
literature to improve the relevance of LBD derived confounders for Alzheimer's Disease (AD). We chose AD for
its high prevalence and relative lack of effective pharmacologic treatment. (A) Compare LBD strategies in a
disease-specific setting. We will test LBD variants using disease-specific literature or with LBD lacking subject-
matter restrictions. (B) Define the ability of robust LBD-informed GCM to validate drug repurposing candidates
for treating AD symptoms. We will test the ability of advanced methods to iteratively resolve hidden latent
confounding, when detected, to improve effect estimates. The fulfillment of these aims will yield new methods
to combine insights from the literature with causal modeling to uncover causal relationships of drug exposures
on adverse events and on beneficial outcomes.
records (EHRs), allow for the identification links between health events, such as drug exposures and side-
effects. Some of these links indicate stable dependencies deemed as causes. Causal insight allows reverse-
engineering disease. If confounding is not addressed, it will be difficult to distinguish causative from correlative
links. Our approach is to identify confounders explicitly. Graphical causal modeling (GCMs) can discover
causal links from data and prior knowledge. GCMs summarize causal links between variables. Automated
selection of variables would allow GCMs to scale and yield more insight from data. Literature-based discovery
(LBD) methods were developed to identify links between concepts in the literature. Advanced methods permit
the search for concepts linked to each other through specific verbs, e.g., “causes”, “treats”. Our hypothesis is
that we can exploit structured knowledge extracted from the literature to inform GCMs. In prior work, we found
that LBD + GCM was better at identifying side-effects in EHR data than traditional methods. Compared to
methods which use solely data, we hypothesize that our method will increase the ability to detect causal
relationships from EHR data. The first aim is to determine the extent to which LBD-informed GCM improves the
identification of causal links for drug safety. We will build LBD-informed GCMs using publicly available
reference datasets for drug safety. These reference datasets contain drug/side-effect pairs for performance
benchmarking. (A) Test the ability of GCM algorithms to identify known causal links solely using data. We will
systematically evaluate GCM algorithms based on their ability to re-discover causal links in a reference
standard. Results will guide our studies on how GCM can be tuned. (B) Determine the effect of adding different
subsets of LBD-derived information to GCMs at identifying drug side-effects. We will build causal models using
increasing numbers confounders. The second aim is to test the ability of LBD built with disease-specific
literature to improve the relevance of LBD derived confounders for Alzheimer's Disease (AD). We chose AD for
its high prevalence and relative lack of effective pharmacologic treatment. (A) Compare LBD strategies in a
disease-specific setting. We will test LBD variants using disease-specific literature or with LBD lacking subject-
matter restrictions. (B) Define the ability of robust LBD-informed GCM to validate drug repurposing candidates
for treating AD symptoms. We will test the ability of advanced methods to iteratively resolve hidden latent
confounding, when detected, to improve effect estimates. The fulfillment of these aims will yield new methods
to combine insights from the literature with causal modeling to uncover causal relationships of drug exposures
on adverse events and on beneficial outcomes.
👤 Principal Investigator(s) (from NIH)
Scott Alexander Malec
🏛️ Recipient Organization (from NIH)
UNIVERSITY OF PITTSBURGH AT PITTSBURGH (PITTSBURGH, PA, UNITED STATES)
📅 Project Dates (from NIH)
Start: 2021-08-01T00:00:00
End: 2023-07-31T00:00:00
End: 2023-07-31T00:00:00
💰 Award Amount (from NIH)
$68,663
📊 Fiscal Year (from NIH)
2022
🏷️ Activity Code (from NIH)
K99
🔢 Project Number (from NIH)
5K99LM013367-02
🔗 Full Project Record (from NIH)
Added
2022-07-20
Last modified
2022-07-20
Source:
DimensionsWizard via Scott Alexander Malec
| ✓ Enriched from NIH
DimensionsWizard via Scott Alexander Malec
| ✓ Enriched from NIH
Using Biomedical Knowledge to Identify Plausible Signals for Pharmacovigilance
2013-09-01 to 2017-08-31 | Grant
United States National Library of Medicine (n/a, US)
GRANT_NUMBER: grant.R01LM011563
Show more detailOrganization identifiers
United States National Library of Medicine: n/a, US
Funding project translated title
Funding project translated title
(en)
Using Biomedical Knowledge to Identify Plausible Signals for Pharmacovigilance
Added
2018-02-06
Last modified
2018-02-06
Source:
DimensionsWizard via Scott Alexander Malec
DimensionsWizard via Scott Alexander Malec
NLM Training Program in Biomedical Informatics & Data Science for Predoctoral and Postdoctoral Fellows
1992-07-01 to 2018-06-30 | Grant
United States National Library of Medicine (n/a, US)
GRANT_NUMBER: grant.T15LM007093
Show more detailOrganization identifiers
United States National Library of Medicine: n/a, US
Funding project translated title
Funding project translated title
(en)
NLM Training Program in Biomedical Informatics & Data Science for Predoctoral and Postdoctoral Fellows
Added
2018-02-06
Last modified
2018-02-06
Source:
DimensionsWizard via Scott Alexander Malec
DimensionsWizard via Scott Alexander Malec
Education and qualifications (4)
University of Pittsburgh: Pittsburgh, PA, US
2018-08-01 to 2023-07-31 |
Postdoctoral Scholar (Department of Biomedical Informatics)
Education
Show more detail
Department
Department of Biomedical Informatics
Added
2018-09-20
Last modified
2024-09-05
Source:
Scott Alexander Malec
University of Texas Health Science Center at Houston: Houston, Texas, US
Department
School of Biomedical Informatics
Added
2018-02-06
Last modified
2018-09-20
Source:
Scott Alexander Malec
Carnegie Mellon University: Pittsburgh, PA, US
Organization identifiers
RINGGOLD:
6612
Carnegie Mellon University : Pittsburgh, PA, US
Department
CMU
Added
2018-02-06
Last modified
2018-02-06
Source:
Scott Alexander Malec
University of Pittsburgh: Pittsburgh, PA, US
2002-01-01 to 2003-12-13 |
MLIS (Department of Library and Information Science)
Education
Show more detail
Organization identifiers
RINGGOLD:
6614
University of Pittsburgh : Pittsburgh, PA, US
Department
Department of Library and Information Science
Added
2018-02-06
Last modified
2018-02-06
Source:
Scott Alexander Malec
CausalKnowledgeTrace: A Novel Computational Framework for Automated Literature-Based Causal Graph Construction and Evidence-Based Variable Selection in Biomedical Research
2026-05-12 | Preprint
Contributors:
Rajesh Upadhayaya; Manjil Pradhan; Vincent Metzger; Scott Alexander Malec
Show more detail
Homepage URL
Contributors
Rajesh Upadhayaya
(Author)
Manjil Pradhan
(Author)
Vincent Metzger
(Author)
Scott Alexander Malec
(Author)
External identifiers
Abstract
Abstract
Background
Variable selection for causal inference from observational biomedical data is challenging, as overlooking confounders or conditioning on colliders leads to biased estimates. While vast causal knowledge exists in biomedical literature, manually extracting this information for principled variable selection is impractical at scale.
Methods
We developed CausalKnowledgeTrace, a Python-based computational framework with Django web interface that systematically leverages structured causal knowledge from the Semantic MEDLINE Database (SemMedDB) to inform variable selection in causal studies. The system implements a six-stage analysis pipeline using NetworkX for graph operations, including graph parsing, basic analysis, comprehensive cycle detection, systematic generic node removal, post-removal analysis, and formal causal inference with bias detection.
Results
Analysis of the hypertension-Alzheimer’s relationship across three degree neighborhoods (1-3) demonstrated systematic scaling of causal complexity: 361-866 variables, 429-1,442 relationships, with graph densities of 0.0033-0.0019. The analysis revealed complex cyclic structures with 54-606 baseline cycles across degree levels. Processing times ranged from 0.3-1.0 seconds for all three degrees, demonstrating computational efficiency for complex biomedical networks. Key confounders identified across all degrees included inflammation, diabetes, insulin resistance, obesity, and ischemia. In the third degree of graph, the pipeline structurally identified 39 confounders, 11 mediators, and 3 colliders from the causal graph. Among the key identified confounders and mediators—including obesity, oxidative stress, ischemia, and vascular diseases—all were found to have strong supporting evidence in established epidemiological and pathophysiological literature.
Conclusions
CausalKnowledgeTrace provides a scalable, evidence-based approach to causal graph construction that systematically identifies confounders and bias structures often missed by conventional approaches. The Python-Django architecture enables both standalone analysis and integration into larger computational workflows, representing a significant advance in computational support for causal inference in biomedical research.
Statement of Significance
Problem or Issue
Selecting proper confounders and variables for causal inference from observational biomedical datasets is challenging and often biased by limited expertise or manual review.
What is Already Known
Existing approaches rely on domain experts, statistical variable screening, or manual construction of causal graphs, but these often overlook literature-documented confounders and complex biases.
What this Paper Adds
This paper introduces an automated, literature-based framework for synthesizing and validating causal graphs, identifying critical variables and complex bias structures, such as M-bias and butterfly bias, with full evidentiary traceability.
Who would benefit from the new knowledge in this paper?
Epidemiologists, biomedical researchers, informaticians, and clinical investigators seeking reliable and transparent causal modeling for observational studies.
Background
Variable selection for causal inference from observational biomedical data is challenging, as overlooking confounders or conditioning on colliders leads to biased estimates. While vast causal knowledge exists in biomedical literature, manually extracting this information for principled variable selection is impractical at scale.
Methods
We developed CausalKnowledgeTrace, a Python-based computational framework with Django web interface that systematically leverages structured causal knowledge from the Semantic MEDLINE Database (SemMedDB) to inform variable selection in causal studies. The system implements a six-stage analysis pipeline using NetworkX for graph operations, including graph parsing, basic analysis, comprehensive cycle detection, systematic generic node removal, post-removal analysis, and formal causal inference with bias detection.
Results
Analysis of the hypertension-Alzheimer’s relationship across three degree neighborhoods (1-3) demonstrated systematic scaling of causal complexity: 361-866 variables, 429-1,442 relationships, with graph densities of 0.0033-0.0019. The analysis revealed complex cyclic structures with 54-606 baseline cycles across degree levels. Processing times ranged from 0.3-1.0 seconds for all three degrees, demonstrating computational efficiency for complex biomedical networks. Key confounders identified across all degrees included inflammation, diabetes, insulin resistance, obesity, and ischemia. In the third degree of graph, the pipeline structurally identified 39 confounders, 11 mediators, and 3 colliders from the causal graph. Among the key identified confounders and mediators—including obesity, oxidative stress, ischemia, and vascular diseases—all were found to have strong supporting evidence in established epidemiological and pathophysiological literature.
Conclusions
CausalKnowledgeTrace provides a scalable, evidence-based approach to causal graph construction that systematically identifies confounders and bias structures often missed by conventional approaches. The Python-Django architecture enables both standalone analysis and integration into larger computational workflows, representing a significant advance in computational support for causal inference in biomedical research.
Statement of Significance
Problem or Issue
Selecting proper confounders and variables for causal inference from observational biomedical datasets is challenging and often biased by limited expertise or manual review.
What is Already Known
Existing approaches rely on domain experts, statistical variable screening, or manual construction of causal graphs, but these often overlook literature-documented confounders and complex biases.
What this Paper Adds
This paper introduces an automated, literature-based framework for synthesizing and validating causal graphs, identifying critical variables and complex bias structures, such as M-bias and butterfly bias, with full evidentiary traceability.
Who would benefit from the new knowledge in this paper?
Epidemiologists, biomedical researchers, informaticians, and clinical investigators seeking reliable and transparent causal modeling for observational studies.
Added
2026-05-25
Last modified
2026-05-25
Source:
Scott Alexander Malec
Scott Alexander Malec
Detecting Uncoded Self-Harm in Veterans' Electronic Health Records Using Positive and Unlabeled Learning: Retrospective Observational Study.
Journal of medical Internet research
2026-04-19 | Journal article
DOI:
10.2196/89071
Contributors:
Kumar P; Viszolay AD; Upadhayaya R; Moomtaheen F; Greer DR (and 12 more)
Show more detail
Homepage URL
Contributors
Kumar P
(Author)
[ORCID: 0000-0002-4981-9020]
Viszolay AD
(Author)
Upadhayaya R
(Author)
[ORCID: 0009-0000-3045-1089]
Moomtaheen F
(Author)
Greer DR
(Author)
Bologa CG
(Author)
Schneider KA
(Author)
Davis SE
(Author)
Matheny ME
(Author)
[ORCID: 0000-0003-3217-4147]
van der Goes D
(Author)
Villarreal G
(Author)
Zhu Y
(Author)
Tohen M
(Author)
Malec SA
(Author)
Yang JJ
(Author)
Fielstein EM
(Author)
Lambert CG
(Author)
External identifiers
PMID:
42001435
DOI:
10.2196/89071
Abstract
BackgroundSuicide and self-harm remain major public health concerns in the United States. Early identification is critical for effective intervention, yet underdiagnosis and undercoding are common across mental health conditions, and only positive cases are typically labeled in healthcare data. As a result, reliable negative examples are missing. Positive and Unlabeled (PU) learning is well-suited to such data, enabling estimation of phenotype prevalence and identification of undiagnosed individuals at elevated risk for self-harm as well as other mental illnesses.ObjectiveTo identify U.S. Veterans whose self-harm events were not explicitly captured through diagnostic codes in electronic health records (EHRs) and estimate the prevalence of ever self-harm cases among Veterans using a novel PU learning algorithm applicable to undetected mental health diagnoses.MethodsWe performed a retrospective observational study using Veterans Health Administration EHRs (from October 1, 1999 to August 31, 2019), selecting a random 25% sample of 1,329,120 Veterans out of 5,316,480 (1,193,563 males and 135,557 females) with at least 2 years of observation. The study cohort comprised 24,625 veterans with coded self-harm and 1,304,495 uncoded for self-harm, with the mean age for coded individuals 38.39 (SD 12.17) and uncoded individuals 48.76 (SD 15.04). We applied our PULSNAR (Positive Unlabeled Learning Selected Not At Random) algorithm to estimate the proportion of individuals with uncoded self-harm. The selected covariates included age at enrollment and the presence or absence of recorded medical conditions, procedures, and clinical observations throughout the observation period. Four experts (raters) independently reviewed charts of 97 uncoded Veterans, each selected from 1% intervals of calibrated PULSNAR probabilities from 0.01 to 0.97. Agreement was assessed among raters, PULSNAR classifications, and consensus review decisions. Post hoc calibration was used to refine prevalence estimates.ResultsOf the 159,049 covariates in the dataset, the XGBoost model within our PULSNAR framework identified 1,302 (0.82%) as informative for classification. Only 1.85% (24,625/1,329,120) of Veterans had diagnostic codes indicating self-harm events, while PULSNAR estimated an overall prevalence of 10.46% (139,026/1,329,120) by identifying an additional ?=8.77% (114,404/1,304,495) of self-harm cases among the uncoded population. Of the 97 chart-reviewed patients, 39 had documented but uncoded self-harm. PULSNAR estimates were post hoc calibrated such that their sum over the 97 cases equaled 39, which resulted in PULSNAR adjusted coded and imputed estimation of 7.91% (105,133/1,329,120). When applied to the 1.3M Veterans, PULSNAR suggests that coded self-harm represents only 23.4% (95% CI: 17.76% to 31.51%) of all documented (coded + notes) self-harm.ConclusionsUnder the Selected Not At Random assumption, PULSNAR provides an innovative and scalable framework for estimating the clinically documented prevalence of mental health conditions and identifying the uncoded individuals with calibrated prediction, without requiring confirmed negative labels. This method offers an alternative to time-consuming chart reviews for detecting likely cases missing structured coding capture. By addressing diagnostic undercoding of mental health conditions in EHRs, this approach has the potential to enhance the estimation of mental health prevalence and support screening, activation of automated clinical decision support, targeted intervention, better resource allocation, and research to improve outcomes in real-world settings.Clinicaltrial
Added
2026-04-27
Last modified
2026-04-27
Source:
Scott Alexander Malec
Scott Alexander Malec
Computational Tools for Target Illumination and Early Stage Drug Discovery: DrugCentral, Badapple, Smartsfilter, TICTAC, TIN-X, TIGA, CKT, and more…
Zenodo
2026-03-08 | Conference poster
Contributors:
Jack Ringer; Bivek Sharma Panthi; Bat Ochir Artur; Jeremiah I Abok; Rajesh Upadhayaya (and 9 more)
Show more detail
Contributors
Jack Ringer
(Author)
[ORCID: 0009-0008-8493-0139]
Bivek Sharma Panthi
(Author)
Bat Ochir Artur
(Author)
Jeremiah I Abok
(Author)
[ORCID: 0000-0003-0119-9181]
Rajesh Upadhayaya
(Author)
[ORCID: 0009-0000-3045-1089]
Manjil M Pradhan
(Author)
Vincent T Metzger
(Author)
[ORCID: 0000-0002-8041-0370]
Scott Alexander Malec
(Author)
[ORCID: 0000-0003-1696-1781]
Ian Watson
(Author)
[ORCID: 0000-0002-3086-9845]
Kerry Fowler
(Author)
Cristian G Bologa
(Author)
[ORCID: 0000-0003-2232-4244]
Tudor I Oprea
(Author)
[ORCID: 0000-0002-6195-6976]
Christophe Gerard Lambert
(Author)
[ORCID: 0000-0003-1994-2893]
Jeremy J Yang
(Author)
[ORCID: 0000-0002-1476-6192]
External identifiers
DOI:
10.5281/zenodo.18905711
Abstract
Presented at the OpenEye - Cadence Molecular Sciences - CUP XXV Meeting, Santa Fe, New Mexico, March 10-12, 2026. New and improved and empowered web apps, APIs, and other computational tools, for target illumination and other early stage drug discovery applications, developed by the Translational Informatics Division, UNM School of Medicine Department of Internal Medicine, with support from many academic, government, and industrial colleagues.
Added
2026-03-08
Last modified
2026-03-08
Source:
DataCite
DataCite
Predicting Alzheimer’s Disease Diagnosis, a Decade or more Years before Onset using the Electronic Health Record and Random Forest Machine Learning Models
2025-11-06 | Preprint
Contributors:
Sanya B. Taneja; Richard D. Boyce; Scott A. Malec; Steven M. Albert; C. Elizabeth Shaaban (and 9 more)
Show more detail
Homepage URL
Contributors
Sanya B. Taneja
(Author)
Richard D. Boyce
(Author)
Scott A. Malec
(Author)
Steven M. Albert
(Author)
C. Elizabeth Shaaban
(Author)
Arthur S. Levine
(Author)
Paul Munro
(Author)
Jiang Bian
(Author)
Jie Xu
(Author)
Demetrius Maraganore
(Author)
Karen Schliep
(Author)
Jonathan C. Silverstein
(Author)
Michelle Kienholz
(Author)
Helmet T. Karim
(Author)
External identifiers
Added
2025-11-07
Last modified
2025-11-12
Source:
Crossref
Crossref
Unsupervised Latent Pattern Analysis for Estimating Type 2 Diabetes Risk in Undiagnosed Populations
2025-10-12 | Conference paper
Contributors:
Praveen Kumar; Vincent T. Metzger; Scott Alexander Malec
Show more detail
Homepage URL
Contributors
Praveen Kumar
(Author)
Vincent T. Metzger
(Author)
Scott Alexander Malec
(Author)
External identifiers
Added
2025-12-10
Last modified
2025-12-10
Source:
Crossref
Crossref
Detecting Opioid Use Disorder in Health Claims Data With Positive Unlabeled Learning
IEEE Journal of Biomedical and Health Informatics
2025-02 | Journal article
Contributors:
Praveen Kumar; Fariha Moomtaheen; Scott A. Malec; Jeremy J. Yang; Cristian G. Bologa (and 9 more)
Show more detail
Homepage URL
Contributors
Praveen Kumar
(Author)
Fariha Moomtaheen
(Author)
Scott A. Malec
(Author)
Jeremy J. Yang
(Author)
Cristian G. Bologa
(Author)
Kristan A Schneider
(Author)
Yiliang Zhu
(Author)
Mauricio Tohen
(Author)
Gerardo Villarreal
(Author)
Douglas J. Perkins
(Author)
Elliot M. Fielstein
(Author)
Sharon E. Davis
(Author)
Michael E. Matheny
(Author)
Christophe G. Lambert
(Author)
External identifiers
Added
2025-02-10
Last modified
2025-02-19
Source:
Crossref
Crossref
Unsupervised Latent Pattern Analysis for Estimating Type 2 Diabetes Risk in Undiagnosed Populations
The 16th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB)
2025 | Conference paper
Contributors:
Praveen Kumar; Vincent T. Metzger; Scott A. Malec
Show more detail
Homepage URL
Contributors
Praveen Kumar
(Author)
Vincent T. Metzger
(Author)
Scott A. Malec
(Author)
External identifiers
Abstract
The global prevalence of diabetes, particularly type 2 diabetes mellitus (T2DM), is rapidly increasing, posing significant health and economic challenges. T2DM not only disrupts blood glucose regulation but also damages vital organs such as the heart, kidneys, eyes, nerves, and blood vessels, leading to substantial morbidity and mortality. In the US alone, the economic burden of diagnosed diabetes exceeded $400 billion in 2022. Early detection of individuals at risk is critical to mitigating these impacts. While machine learning approaches for T2DM prediction are increasingly adopted, many rely on supervised learning, which is often limited by the lack of confirmed negative cases. To address this limitation, we propose a novel unsupervised framework that integrates Non-negative Matrix Factorization (NMF) with statistical techniques to identify individuals at risk of developing T2DM. Our method identifies latent patterns of multimorbidity and polypharmacy among diagnosed T2DM patients and applies these patterns to estimate the T2DM risk in undiagnosed individuals. By leveraging data-driven insights from comorbidity and medication usage, our approach provides an interpretable and scalable solution that can assist healthcare providers in implementing timely interventions, ultimately improving patient outcomes and potentially reducing the future health and economic burden of T2DM.
Added
2025-10-01
Last modified
2025-10-07
Source:
Scott Alexander Malec
Scott Alexander Malec
An open source knowledge graph ecosystem for the life sciences.
Scientific data
2024-04-11 | Journal article
Contributors:
Callahan TJ; Tripodi IJ; Stefanski AL; Cappelletti L; Taneja SB (and 27 more)
Show more detail
Homepage URL
Contributors
Callahan TJ
(Author)
[ORCID: 0000-0002-8169-9049]
Tripodi IJ
(Author)
Stefanski AL
(Author)
Cappelletti L
(Author)
Taneja SB
(Author)
[ORCID: 0000-0003-1707-1617]
Wyrwa JM
(Author)
Casiraghi E
(Author)
[ORCID: 0000-0003-2024-7572]
Matentzoglu NA
(Author)
[ORCID: 0000-0002-7356-1779]
Reese J
(Author)
Silverstein JC
(Author)
[ORCID: 0000-0002-9252-6039]
Hoyt CT
(Author)
[ORCID: 0000-0003-4423-4370]
Boyce RD
(Author)
Malec SA
(Author)
Unni DR
(Author)
[ORCID: 0000-0002-3583-7340]
Joachimiak MP
(Author)
Robinson PN
(Author)
[ORCID: 0000-0002-0736-9199]
Mungall CJ
(Author)
Cavalleri E
(Author)
[ORCID: 0000-0003-1973-5712]
Fontana T
(Author)
Valentini G
(Author)
[ORCID: 0000-0002-5694-3919]
Mesiti M
(Author)
[ORCID: 0000-0001-5701-0080]
Gillenwater LA
(Author)
Santangelo B
(Author)
Vasilevsky NA
(Author)
[ORCID: 0000-0001-5208-3432]
Hoehndorf R
(Author)
[ORCID: 0000-0001-8149-5890]
Bennett TD
(Author)
Ryan PB
(Author)
Hripcsak G
(Author)
Kahn MG
(Author)
[ORCID: 0000-0003-4786-6875]
Bada M
(Author)
Baumgartner WA Jr
(Author)
Hunter LE
(Author)
External identifiers
Abstract
Translational research requires data at multiple scales of biological organization. Advancements in sequencing and multi-omics technologies have increased the availability of these data, but researchers face significant integration challenges. Knowledge graphs (KGs) are used to model complex phenomena, and methods exist to construct them automatically. However, tackling complex biomedical integration problems requires flexibility in the way knowledge is modeled. Moreover, existing KG construction methods provide robust tooling at the cost of fixed or limited choices among knowledge representation models. PheKnowLator (Phenotype Knowledge Translator) is a semantic ecosystem for automating the FAIR (Findable, Accessible, Interoperable, and Reusable) construction of ontologically grounded KGs with fully customizable knowledge representation. The ecosystem includes KG construction resources (e.g., data preparation APIs), analysis tools (e.g., SPARQL endpoint resources and abstraction algorithms), and benchmarks (e.g., prebuilt KGs). We evaluated the ecosystem by systematically comparing it to existing open-source KG construction methods and by analyzing its computational performance when used to construct 12 different large-scale KGs. With flexible knowledge representation, PheKnowLator enables fully customizable KGs without compromising performance or usability.
Added
2025-11-17
Last modified
2025-11-17
Source:
Scott Alexander Malec
Scott Alexander Malec
Causal feature selection using a knowledge graph combining structured knowledge from the biomedical literature and ontologies: A use case studying depression as a risk factor for Alzheimer’s disease
Journal of Biomedical Informatics
2023-06 | Journal article
Contributors:
Scott A. Malec; Sanya B. Taneja; Steven M. Albert; C. Elizabeth Shaaban; Helmet T. Karim (and 4 more)
Show more detail
Homepage URL
Contributors
Scott A. Malec
(Author)
Sanya B. Taneja
(Author)
Steven M. Albert
(Author)
C. Elizabeth Shaaban
(Author)
Helmet T. Karim
(Author)
Arthur S. Levine
(Author)
Paul Munro
(Author)
Tiffany J. Callahan
(Author)
Richard D. Boyce
(Author)
External identifiers
Added
2023-05-25
Last modified
2025-02-19
Source:
Crossref
Crossref
An Open-Source Knowledge Graph Ecosystem for the Life Sciences
arXiv
2023 | Preprint
Contributors:
Tiffany J. Callahan; Ignacio J. Tripodi; Adrianne L. Stefanski; Luca Cappelletti; Sanya B. Taneja (and 27 more)
Show more detail
Homepage URL
Contributors
Tiffany J. Callahan
(Author)
Ignacio J. Tripodi
(Author)
Adrianne L. Stefanski
(Author)
Luca Cappelletti
(Author)
Sanya B. Taneja
(Author)
Jordan M. Wyrwa
(Author)
Elena Casiraghi
(Author)
Nicolas A. Matentzoglu
(Author)
Justin Reese
(Author)
Jonathan C. Silverstein
(Author)
Charles Tapley Hoyt
(Author)
Richard D. Boyce
(Author)
Scott A. Malec
(Author)
Deepak R. Unni
(Author)
Marcin P. Joachimiak
(Author)
Peter N. Robinson
(Author)
Christopher J. Mungall
(Author)
Emanuele Cavalleri
(Author)
Tommaso Fontana
(Author)
Giorgio Valentini
(Author)
Marco Mesiti
(Author)
Lucas A. Gillenwater
(Author)
Brook Santangelo
(Author)
Nicole A. Vasilevsky
(Author)
Robert Hoehndorf
(Author)
Tellen D. Bennett
(Author)
Patrick B. Ryan
(Author)
George Hripcsak
(Author)
Michael G. Kahn
(Author)
Michael Bada
(Author)
William A. Baumgartner
(Author)
Lawrence E. Hunter
(Author)
External identifiers
Abstract
Translational research requires data at multiple scales of biological organization. Advancements in sequencing and multi-omics technologies have increased the availability of these data but researchers face significant integration challenges. Knowledge graphs (KGs) are used to model complex phenomena, and methods exist to automatically construct them. However, tackling complex biomedical integration problems requires flexibility in the way knowledge is modeled. Moreover, existing KG construction methods provide robust tooling at the cost of fixed or limited choices among knowledge representation models. PheKnowLator (Phenotype Knowledge Translator) is a semantic ecosystem for automating the FAIR (Findable, Accessible, Interoperable, and Reusable) construction of ontologically grounded KGs with fully customizable knowledge representation. The ecosystem includes KG construction resources (e.g., data preparation APIs), analysis tools (e.g., SPARQL endpoints and abstraction algorithms), and benchmarks (e.g., prebuilt KGs and embeddings). We evaluate the ecosystem by surveying open-source KG construction methods and analyzing its computational performance when constructing 12 large-scale KGs. With flexible knowledge representation, PheKnowLator enables fully customizable KGs without compromising performance or usability.
Added
2023-11-16
Last modified
2025-02-19
Source:
Scott Alexander Malec
Scott Alexander Malec
Causal feature selection using a knowledge graph combining structured knowledge from the biomedical literature and ontologies: a use case studying depression as a risk factor for Alzheimer's disease
2022-07-20 | Preprint
Contributors:
Scott Alexander Malec; Sanya B Taneja; Steven M Albert; C. Elizabeth Shaaban; Helmet T Karim (and 4 more)
Show more detail
Homepage URL
Contributors
Scott Alexander Malec
(Author)
Sanya B Taneja
(Author)
Steven M Albert
(Author)
C. Elizabeth Shaaban
(Author)
Helmet T Karim
(Author)
Art S Levine
(Author)
Paul Wesley Munro
(Author)
Tiffany J Callahan
(Author)
Richard David Boyce
(Author)
External identifiers
Added
2022-07-20
Last modified
2025-02-19
Source:
Crossref
Crossref
Does clinical data capture modifiable midlife risk factors for Alzheimer’s disease?
Alzheimer's & Dementia
2021-12 | Journal article
DOI:
10.1002/alz.055756
Contributors:
C Elizabeth Shaaban; Sanya B Taneja; Kailyn F Witonsky; Scott A Malec; Helmet T Karim (and 5 more)
Show more detail
Homepage URL
Contributors
C Elizabeth Shaaban
(Author)
Sanya B Taneja
(Author)
Kailyn F Witonsky
(Author)
Scott A Malec
(Author)
Helmet T Karim
(Author)
Sheila Pratt
(Author)
Arthur S Levine
(Author)
Paul Munro
(Author)
Richard D Boyce
(Author)
Steven M Albert
(Author)
External identifiers
Added
2023-04-12
Last modified
2025-02-19
Source:
Scott Alexander Malec
Scott Alexander Malec
Using computable knowledge mined from the literature to elucidate confounders for EHR-based pharmacovigilance
2020-07-10 | Other
Contributors:
Scott A. Malec; Elmer V. Bernstam; Peng Wei; Richard D. Boyce; Trevor Cohen
Show more detail
Homepage URL
Contributors
Scott A. Malec
(Author)
Elmer V. Bernstam
(Author)
Peng Wei
(Author)
Richard D. Boyce
(Author)
Trevor Cohen
(Author)
External identifiers
Added
2020-07-14
Last modified
2025-02-19
Source:
Crossref
Crossref
Exploring Novel Computable Knowledge in Structured Drug Product Labels.
AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science
2020-05 | Journal article
Contributors:
Malec SA; Boyce RD
Show more detail
Homepage URL
Contributors
Malec SA
(Author)
Boyce RD
(Author)
External identifiers
PMID:
32477661
PMC:
PMC7233092
Added
2020-07-08
Last modified
2025-02-19
Source:
Europe PubMed Central
Europe PubMed Central
Literature-Based Discovery of Confounding in Observational Clinical Data.
AMIA Annual Symposium Proceedings
2016 | Journal article
Contributors:
Malec SA; Wei P; Xu H; Bernstam EV; Myneni S (and 1 more)
Show more detail
Homepage URL
Contributors
Malec SA
(Author)
Wei P
(Author)
Xu H
(Author)
Bernstam EV
(Author)
Myneni S
(Author)
Cohen T
(Author)
External identifiers
PMID:
28269951
PMC:
PMC5333204
Added
2018-02-06
Last modified
2025-02-19
Source:
Europe PubMed Central
Europe PubMed Central
Propp Revisited: Integration of Linguistic Markup into Structured Content Descriptors of Tales
Added
2018-02-07
Last modified
2025-02-19
Source:
Scott Alexander Malec
Scott Alexander Malec
Integration of Linguistic Markup into Semantic Models of Folk Narratives: The Fairy Tale Use Case.
Unpublished
2010 | Conference paper
Contributors:
Piroska Lendvai; Thierry Declerck; Sándor Darányi; Pablo Gervás; Raquel Hervás (and 2 more)
Show more detail
Contributors
Piroska Lendvai
(Author)
Thierry Declerck
(Author)
Sándor Darányi
(Author)
Pablo Gervás
(Author)
Raquel Hervás
(Author)
Scott Malec
(Author)
Federico Peinado
(Author)
External identifiers
DOI:
10.13140/2.1.2365.2801
Added
2018-02-06
Last modified
2025-10-09
Source:
DataCite
DataCite
AutoPropp: Toward the Automatic Markup, Classification, and Annotation of Russian Magic Tales
Other
Show more detail
Added
2018-02-06
Last modified
2025-02-19
Source:
Scott Alexander Malec
Scott Alexander Malec