Long-term reliability of intracortical microelectrodes remains a challenge for increased acceptance and deployment. There are conflicting reports comparing measurements associated with recording quality with postmortem histology, in attempts to better understand failure of intracortical microelectrodes (IMEs). Our group has recently introduced the assessment of motor behavior tasks as another metric to evaluate the effects of IME implantation. We hypothesized that adding the third dimension to our analysis, functional behavior testing, could provide substantial insight on the health of the tissue, success of surgery/implantation, and the long-term performance of the implanted device. Here we present our novel analysis scheme including: (1) the use of numerical formal concept analysis (nFCA) and (2) a regression analysis utilizing modern model/variable selection. The analyses found complimentary relationships between the variables. The histological variables for glial cell activation had associations between each other, as well as the neuronal density around the electrode interface. The neuronal density had associations to the electrophysiological recordings and some of the motor behavior metrics analyzed. The novel analyses presented herein describe a valuable tool that can be utilized to assess and understand relationships between diverse variables being investigated. These models can be applied to a wide range of ongoing investigations utilizing various devices and therapeutics.

The increasing use of Real-World Data (RWD) in clinical research is critical for evidence-based decision making but presents challenges to data analytics. Unlike in Randomized Controlled Trials (RCTs), missingness in RWD occurs often and can include complex patterns which may or may not be Missing at Random (MAR). Informative absence (Missing Not at Random, or MNAR) occurs when the absence of data itself is a clinical signal. Applications of ad-hoc methods, or even popular “universal” methods, can lead to biased inferences when these varied data problems are not appropriately addressed. This paper introduces a multi-step algorithm known as D ivide A nd C onquer, r ejoin (DACr, pronounced as “DACK-er”) for handling missing data in RWD. We illustrate the algorithm’s application using a large-scale, high-dimensional heart transplant dataset (from p = 1107 to 527 after initial screening) from the Scientific Registry of Transplant Recipients (SRTR). This “divide and conquer” idea handles variables based on data type, the proportion, and type of missingness. It also allows for the targeted application of justified data handling techniques: appropriate imputation method is used for variables assumed to be MAR, while variables with plausible MNAR mechanisms are recoded as categorical with an extra level of “missing” to preserve their informative signal. The proposed algorithm provides a systematic treatment for practical researchers.

Feature selection from a large number of covariates (aka features) in a regression analysis remains a challenge in data science, especially in terms of its potential of scaling to ever-enlarging data and finding a group of scientifically meaningful features. For example, to develop new, responsive drug targets for ovarian cancer, the actual false discovery rate (FDR) of a practical feature selection procedure must also match the target FDR. The popular approach to feature selection, when true features are sparse, is to use a penalized likelihood or a shrinkage estimation, such as a LASSO, SCAD, Elastic Net, or MCP procedure (call them benchmark procedures). We present a different approach using a new subsampling method, called the Subsampling Winner algorithm (SWA). The central idea of SWA is analogous to that used for the selection of US national merit scholars. SWA uses a "base procedure" to analyze each of the subsamples, computes the scores of all features according to the performance of each feature from all subsample analyses, obtains the "semifinalist" based on the resulting scores, and then determines the "finalists," i.e., the most important features. Due to its subsampling nature, SWA can scale to data of any dimension in principle. The SWA also has the best-controlled actual FDR in comparison with the benchmark procedures and the randomForest, while having a competitive true-feature discovery rate. We also suggest practical add-on strategies to SWA with or without a penalized benchmark procedure to further assure the chance of "true" discovery. Our application of SWA to the ovarian serous cystadenocarcinoma specimens from the Broad Institute revealed functionally important genes and pathways, which we verified by additional genomics tools. This second-stage investigation is essential in the current discussion of the proper use of P-values.