<i>In silico</i> analysis of TRPM4 variants of unknown clinical significance

by Svetlana I. Tarnovskaya, Anna A. Kostareva, Boris S. Zhorov

Background

TRPM4 is a calcium-activated channel that selectively permeates monovalent cations. Genetic variants of the channel in cardiomyocytes are associated with various heart disorders, such as progressive familial heart block and Brugada syndrome. About97% of all known TRPM4 missense variants are classified as variants of unknown clinical significance (VUSs). The very large number of VUSs is a serious problem in diagnostics and treatment of inherited heart diseases.

Methods and results

We collected 233 benign or pathogenic missense variants in the superfamily of TRP channels from databases ClinVar, Humsavar and Ensembl Variation to compare performance of 22 algorithms that predict damaging variants. We found that ClinPred is the best-performing tool for TRP channels. We also used the paralogue annotation method to identify disease variants across the TRP family. In the set of 565 VUSs of hTRPM4, ClinPred predicted pathogenicity of 299 variants. Among these, 12 variants are also categorized as LP/P variants in at least one paralogue of hTRPM4. We further used the cryo-EM structure of hTRPM4 to find scores of contact pairs between parental (wild type) residues of VUSs for which ClinPred predicts a high probability of pathogenicity of variants for both contact partners. We propose that 68 respective missense VUSs are also likely pathogenic variants.

Conclusions

ClinPred outperformed other in-silico tools in predicting damaging variants of TRP channels. ClinPred, the paralogue annotation method, and analysis of residue contacts the hTRPM4 cryo-EM structure collectively suggest pathogenicity of 80 TRPM4 VUSs.

Robust cardiac segmentation corrected with heuristics

by Alan Cervantes-Guzmán, Kyle McPherson, Jimena Olveres, Carlos Francisco Moreno-García, Fabián Torres Robles, Eyad Elyan, Boris Escalante-Ramírez

Cardiovascular diseases related to the right side of the heart, such as Pulmonary Hypertension, are some of the leading causes of death among the Mexican (and worldwide) population. To avoid invasive techniques such as catheterizing the heart, improving the segmenting performance of medical echocardiographic systems can be an option to early detect diseases related to the right-side of the heart. While current medical imaging systems perform well segmenting automatically the left side of the heart, they typically struggle segmenting the right-side cavities. This paper presents a robust cardiac segmentation algorithm based on the popular U-NET architecture capable of accurately segmenting the four cavities with a reduced training dataset. Moreover, we propose two additional steps to improve the quality of the results in our machine learning model, 1) a segmentation algorithm capable of accurately detecting cone shapes (as it has been trained and refined with multiple data sources) and 2) a post-processing step which refines the shape and contours of the segmentation based on heuristics provided by the clinicians. Our results demonstrate that the proposed techniques achieve segmentation accuracy comparable to state-of-the-art methods in datasets commonly used for this practice, as well as in datasets compiled by our medical team. Furthermore, we tested the validity of the post-processing correction step within the same sequence of images and demonstrated its consistency with manual segmentations performed by clinicians.