MSTPublications: August 2021
Machine Learning Prediction of Kidney Stone Composition Using Electronic Health Record-Derived Features. Abraham A, Kavoussi N, Sui W, Bejan C, Capra JA, Hsi R. J Endourol. 2021 Jul 27. doi: 10.1089/end.2021.0211. Online ahead of print.
Aims: Noninvasive prediction of kidney stone composition could direct dietary and pharmacologic preventative treatment without stone analysis. We aimed to assess the accuracy of machine learning models in predicting kidney stone composition using variables extracted from the electronic health record (EHR).
Materials and methods: We identified kidney stone patients (n=1,296) with both stone composition and 24-hour (24H) urine testing. We trained machine learning models (XGBoost [XG] and logistic regression [LR]) to predict stone composition using 24H urine data and EHR-derived demographic and comorbidity data. Models predicted either binary (calcium vs. non-calcium stone) or multiclass (calcium oxalate, uric acid, hydroxyapatite, or other) stone types. We evaluated performance using area under the receiver operating curve (ROC-AUC) and accuracy and identified predictors for each task.
Results: For discriminating binary stone composition, XG outperformed LR with higher accuracy (91% vs. 71%) with ROC-AUC of 0.80 for both models. Top predictors used by these models were supersaturations of uric acid and calcium phosphate, and urinary ammonium. For multiclass classification, LR outperformed XG with higher accuracy (0.64 vs. 0.56) and ROC-AUC (0.79 vs. 0.59), and urine pH had the highest predictive utility. Overall, 24H urine analyte data contributed more to the models’ predictions of stone composition than EHR-derived variables.
Conclusion: Machine learning models can predict calcium stone composition. LR outperforms XG in multiclass stone classification. Demographic and comorbidity data are predictive of stone composition; however, including 24H urine data improves performance. Further optimization of performance could lead to earlier, directed medical therapy for kidney stone patients.
Tissue-specific expression of p73 and p63 isoforms in human tissues. Marshall CB, Beeler JS, Lehmann BD, Gonzalez-Ericsson P, Sanchez V, Sanders ME, Boyd KL, Pietenpol JA. Cell Death Dis. 2021 Jul 27;12(8):745. doi: 10.1038/s41419-021-04017-8.
p73 and p63 are members of the p53 family that exhibit overlapping and distinct functions in development and homeostasis. The evaluation of p73 and p63 isoform expression across human tissue can provide greater insight to the functional interactions between family members. We determined the mRNA isoform expression patterns of TP73 and TP63 across a panel of 36 human tissues and protein expression within the highest-expressing tissues. TP73 and TP63 expression significantly correlated across tissues. In tissues with concurrent mRNA expression, nuclear co-expression of both proteins was observed in a majority of cells. Using GTEx data, we quantified p73 and p63 isoform expression in human tissue and identified that the α-isoforms of TP73 and TP63 were the predominant isoform expressed in nearly all tissues. Further, we identified a previously unreported p73 mRNA product encoded by exons 4 to 14. In sum, these data provide the most comprehensive tissue-specific atlas of p73 and p63 protein and mRNA expression patterns in human and murine samples, indicating coordinate expression of these transcription factors in the majority of tissues in which they are expressed.
Joint cortical surface and structural connectivity analysis of Alzheimer’s Disease. Cai LY, Kerley CI, Yu C, Aboud KS, Beason-Held LL, Shafer AT, Resnick SM, Jordan LC, Anderson AW, Schilling KG, Lyu I, Landman BA. Proc SPIE Int Soc Opt Eng. 2021;11596:1159630. doi: 10.1117/12.2580956. Epub 2021 Feb 15.
Prior neuroimaging studies have demonstrated isolated structural and connectivity changes in the brain due to Alzheimer’s Disease (AD). However, how these changes relate to each other is not well understood. We present a preliminary study to begin to fill this gap by leveraging joint independent component analysis (jICA). We explore how jICA performs in an analysis of T1 and diffusion weighted MRI by characterizing the joint changes of complex cortical surface and structural connectivity metrics in AD in subjects from the Baltimore Longitudinal Study of Aging. We calculate 588 region-based cortical metrics and 4,753 fractional anisotropy-based connectivity metrics and project them into a low-dimensional manifold with principal component analysis. We perform jICA on the manifold and subsequently backproject the independent components to the original data space. We demonstrate component stability with 3-fold cross validation and find differential component loadings between 776 cognitively unimpaired control subjects and 23 with AD that generalizes across folds. In addition, we perform the same analysis on the surface and connectivity metrics separately and find that the joint approach identifies both novel and similar components to the separate approaches. To illustrate the joint approach’s primary utility, we provide an example hypothesis for how surface and connectivity components may vary together with AD. These preliminary results suggest jointly varying independent cortical surface and structural connectivity components can be consistently extracted from MRI data and provide a data-driven way for generating novel hypotheses about AD that may not be captured by separate analyses.
Joint analysis of structural connectivity and cortical surface features: correlates with mild traumatic brain injury. Kerley CI, Cai LY, Yu C, Crawford LM, Elenberger JM, Singh ES, Schilling KG, Aboud KS, Landman BA, Rex TS. Proc SPIE Int Soc Opt Eng. 2021;11596:115960R. doi: 10.1117/12.2580902. Epub 2021 Feb 15.
Fiber tractography bundle segmentation depends on scanner effects, vendor effects, acquisition resolution, diffusion sampling scheme, diffusion sensitization, and bundle segmentation workflow. Schilling KG, Tax CMW, Rheault F, Hansen C, Yang Q, Yeh FC, Cai L, Anderson AW, Landman BA. Neuroimage. 2021 Aug 3:118451. doi: 10.1016/j.neuroimage.2021.118451. Online ahead of print.
Urinary tract infections in cystic fibrosis patients.Reasoner SA, Enriquez KT, Abelson B, Scaglione S, Schneier B, O’Connor MG, Van Horn G, Hadjifrangiskou M. J Cyst Fibros. 2021 Jul 27:S1569-1993(21)01303-5. doi: 10.1016/j.jcf.2021.07.005. Online ahead of print.
Improved understanding of non-respiratory infections in cystic fibrosis (CF) patients will be vital to sustaining the increased life span of these patients. To date, there has not been a published report of urinary tract infections (UTIs) in CF patients. We performed a retrospective chart review at a major academic medical center during 2010-2020 to determine the features of UTIs in 826 CF patients. We identified 108 UTI episodes during this period. Diabetes, distal intestinal obstruction syndrome (DIOS), and nephrolithiasis were correlated with increased risk of UTIs. UTIs in CF patients were less likely to be caused by Gram-negative rods compared to non-CF patients and more likely to be caused by Enterococcus faecalis. The unique features of UTIs in CF patients highlight the importance of investigating non-respiratory infections to ensure appropriate treatment.
Measuring Depression in Autistic Adults: Psychometric Validation of the Beck Depression Inventory-II. Williams ZJ, Everaert J, Gotham KO. Assessment. 2021 Apr;28(3):858-876. doi: 10.1177/1073191120952889. Epub 2020 Aug 29.
Depressive disorders are common in autistic adults, but few studies have examined the extent to which common depression questionnaires are psychometrically appropriate for use in this population. Using item response theory, this study examined the psychometric properties of the Beck Depression Inventory-II (BDI-II) in a sample of 947 autistic adults. BDI-II latent trait scores exhibited strong reliability, construct validity, and moderate ability to discriminate between depressed and nondepressed adults on the autism spectrum (area under the receiver operating characteristic curve = 0.796 [0.763, 0.826], sensitivity = 0.820 [0.785, 0.852], specificity = 0.653 [0.601, 0.699]). These results collectively indicate that the BDI-II is a valid measure of depressive symptoms in autistic adults, appropriate for quantifying depression severity in research studies or screening for depressive disorders in clinical settings. A free online score calculator has been created to facilitate the use of BDI-II latent trait scores for clinical and research applications (available at https://asdmeasures.shinyapps.io/bdi_score/).
Improving the measurement of alexithymia in autistic adults: a psychometric investigation of the 20-item Toronto Alexithymia Scale and generation of a general alexithymia factor score using item response theory. Williams ZJ, Gotham KO. Mol Autism. 2021 Aug 10;12(1):56. doi: 10.1186/s13229-021-00463-5.
Background: Alexithymia, a personality trait characterized by difficulties interpreting emotional states, is commonly elevated in autistic adults, and a growing body of literature suggests that this trait underlies several cognitive and emotional differences previously attributed to autism. Although questionnaires such as the 20-item Toronto Alexithymia Scale (TAS-20) are frequently used to measure alexithymia in the autistic population, few studies have investigated the psychometric properties of these questionnaires in autistic adults, including whether differential item functioning (I-DIF) exists between autistic and general population adults.
Methods: This study is a revised version of a previous article that was retracted due to copyright concerns (Williams and Gotham in Mol Autism 12:1-40). We conducted an in-depth psychometric analysis of the TAS-20 in a large sample of 743 cognitively able autistic adults recruited from the Simons Foundation SPARK participant pool and 721 general population controls enrolled in a large international psychological study. The factor structure of the TAS-20 was examined using confirmatory factor analysis, and item response theory was used to generate a subset of the items that were strong indicators of a “general alexithymia” factor. Correlations between alexithymia and other clinical outcomes were used to assess the nomological validity of the new alexithymia score in the SPARK sample.
Results: The TAS-20 did not exhibit adequate model fit in either the autistic or general population samples. Empirically driven item reduction was undertaken, resulting in an 8-item general alexithymia factor score (GAFS-8, with “TAS” no longer referenced due to copyright) with sound psychometric properties and practically ignorable I-DIF between diagnostic groups. Correlational analyses indicated that GAFS-8 scores, as derived from the TAS-20, meaningfully predict autistic trait levels, repetitive behaviors, and depression symptoms, even after controlling for trait neuroticism. The GAFS-8 also presented no meaningful decrement in nomological validity over the full TAS-20 in autistic participants.
Limitations: Limitations of the current study include a sample of autistic adults that was majority female, later diagnosed, and well educated; clinical and control groups drawn from different studies with variable measures; only 16 of the TAS-20 items being administered to the non-autistic sample; and an inability to test several other important psychometric characteristics of the GAFS-8, including sensitivity to change and I-DIF across multiple administrations.
Conclusions: These results indicate the potential of the GAFS-8 to robustly measure alexithymia in both autistic and non-autistic adults. A free online score calculator has been created to facilitate the use of norm-referenced GAFS-8 latent trait scores in research applications (available at https://asdmeasures.shinyapps.io/alexithymia ).
Development and Validation of an Electronic Medical Record Algorithm to Identify Phenotypes of Rotator Cuff Tear. Gao C, Fan R, Ayers GD, Giri A, Harris K, Atreya R, Teixeira PL, Jain NB. PM R. 2020 Nov;12(11):1099-1105. doi: 10.1002/pmrj.12367. Epub 2020 Apr 29.
Gonzalez-Ericsson PI, Wulfkhule JD, Gallagher RI, Sun X, Axelrod ML, Sheng Q, Luo N, Gomez H, Sanchez V, Sanders M, Pusztai L, Petricoin EF, Blenman KR, Balko JM, Trial Team IS. Clin Cancer Res. 2021 Jul 27:clincanres.CCR-21-0607-A.2021. doi: 10.1158/1078-0432.CCR-21-0607. Online ahead of print.
Computationally Designed Cyclic Peptides Derived from an Antibody Loop Increase Breadth of Binding for Influenza Variants. Sevy AM, Gilchuk IM, Brown BP, Bozhanova NG, Nargi R, Jensen M, Meiler J, Crowe JE Jr. Structure. 2020 Oct 6;28(10):1114-1123.e4. doi: 10.1016/j.str.2020.04.005. Epub 2020 Jun 30.
Top-Down Fabricated microPlates for Prolonged, Intra-articular Matrix Metalloproteinase 13 siRNA Nanocarrier Delivery to Reduce Post-traumatic Osteoarthritis. Bedingfield SK, Colazo JM, Di Francesco M, Yu F, Liu DD, Di Francesco V, Himmel LE, Gupta MK, Cho H, Hasty KA, Decuzzi P, Duvall CL. ACS Nano. 2021 Aug 19. doi: 10.1021/acsnano.1c04005. Online ahead of print.
Kappa opioid receptor modulation of excitatory drive onto nucleus accumbens fast-spiking interneurons. Coleman BC, Manz KM, Grueter BA. Neuropsychopharmacology. 2021 Aug 16. doi: 10.1038/s41386-021-01146-8. Online ahead of print.
Metabolic pre-conditioning in CD4 T cells restores inducible immune tolerance in lupus prone mice. Wilson CS, Stocks BT, Hoopes EM, Rhoads JP, McNew KL, Major AS, Moore DJ. JCI Insight. 2021 Aug 17:143245. doi: 10.1172/jci.insight.143245. Online ahead of print.
Inhibition of Histone Deacetylases 1, 2, and 3 Enhances Clearance of Cholesterol Accumulation in Niemann-Pick C1 Fibroblasts. Dana L. Cruz, Nina Pipalia, Shu Mao, Deepti Gadi, Gang Liu, Michael Grigalunas, Matthew O’Neill, Taylor R. Quinn, Andi Kipper, Andreas Ekebergh, Alexander Dimmling, Carlos Gartner, Bruce J. Melancon, Florence F. Wagner, Edward Holson, Paul Helquist, Olaf Wiest, and Frederick R. Maxfield. ACS Pharmacology and Translational Science. 2021, 4, 1136.
TRPM7 is a critical regulator of pancreatic endocrine development and high-fat diet-induced β-cell proliferation. Altman MK, Schaub CM, Dickerson MT, Zaborska KE, Dadi PK, Nakhe AY, Graff SM, Galletta TJ, Amarnath G, Thorson AS, Gu G, Jacobson DA. Development. 2021 Aug 4:dev.194928. doi: 10.1242/dev.194928. Online ahead of print.
Combinatorial transcription factor profiles predict mature and functional human islet α and β cells. Shrestha S, Saunders DC, Walker JT, Camunas-Soler J, Dai XQ, Haliyur R, Aramandla R, Poffenberger G, Prasad N, Bottino R, Stein R, Cartailler JP, Parker SC, MacDonald PE, Levy SE, Powers AC, Brissova M. JCI Insight. 2021 Aug 24:151621. doi: 10.1172/jci.insight.151621. Online ahead of print.
Measurement of Osteoblast Cytotoxicity Induced by S. aureus Secreted Toxins. Ford CA, Cassat JE. Methods Mol Biol. 2021;2341:141-152. doi: 10.1007/978-1-0716-1550-8_17.