The traditional framework for feature selection treats all features as costing the same amount. However, in reality, a scientist often has considerable discretion regarding which variables to measure, and the decision involves a tradeoff between model accuracy and cost (where cost can refer to money, time, difficulty or intrusiveness). In particular, unnecessarily including an expensive feature in a model is worse than unnecessarily including a cheap feature. We propose a procedure, which we call cheap knockoffs, for performing feature selection in a cost-conscious manner. The key idea behind our method is to force higher cost features to compete with more knockoffs than cheaper features. We derive an upper bound on the weighted false discovery proportion associated with this procedure, which corresponds to the fraction of the feature cost that is wasted on unimportant features. We prove that this bound holds simultaneously with high probability over a path of selected variable sets of increasing size. A user may thus select a set of features based, for example, on the overall budget, while knowing that no more than a particular fraction of feature cost is wasted. We investigate, through simulation and a biomedical application, the practical importance of incorporating cost considerations into the feature selection process.

We consider the problem of estimating the parameters in a pairwise graphical model in which the distribution of each node, conditioned on the others, may have a different exponential family form. We identify restrictions on the parameter space required for the existence of a well-defined joint density, and establish the consistency of the neighbourhood selection approach for graph reconstruction in high dimensions when the true underlying graph is sparse. Motivated by our theoretical results, we investigate the selection of edges between nodes whose conditional distributions take different parametric forms, and show that efficiency can be gained if edge estimates obtained from the regressions of particular nodes are used to reconstruct the graph. These results are illustrated with examples of Gaussian, Bernoulli, Poisson and exponential distributions. Our theoretical findings are corroborated by evidence from simulation studies.

BACKGROUND: Pathologists analyze biopsy material at both the cellular and structural level to determine diagnosis and cancer stage. Mitotic figures are surrogate biomarkers of cellular proliferation that can provide prognostic information; thus, their precise detection is an important factor for clinical care. Convolutional Neural Networks (CNNs) have shown remarkable performance on several recognition tasks. Utilizing CNNs for mitosis classification may aid pathologists to improve the detection accuracy. METHODS: We studied two state-of-the-art CNN-based models, ESPNet and DenseNet, for mitosis classification on six whole slide images of skin biopsies and compared their quantitative performance in terms of sensitivity, specificity, and F-score. We used raw RGB images of mitosis and non-mitosis samples with their corresponding labels as training input. In order to compare with other work, we studied the performance of these classifiers and two other architectures, ResNet and ShuffleNet, on the publicly available MITOS breast biopsy dataset and compared the performance of all four in terms of precision, recall, and F-score (which are standard for this data set), architecture, training time and inference time. RESULTS: The ESPNet and DenseNet results on our primary melanoma dataset had a sensitivity of 0.976 and 0.968, and a specificity of 0.987 and 0.995, respectively, with F-scores of .968 and .976, respectively. On the MITOS dataset, ESPNet and DenseNet showed a sensitivity of 0.866 and 0.916, and a specificity of 0.973 and 0.980, respectively. The MITOS results using DenseNet had a precision of 0.939, recall of 0.916, and F-score of 0.927. The best published result on MITOS (Saha et al. 2018) reported precision of 0.92, recall of 0.88, and F-score of 0.90. In our architecture comparisons on MITOS, we found that DenseNet beats the others in terms of F-Score (DenseNet 0.927, ESPNet 0.890, ResNet 0.865, ShuffleNet 0.847) and especially Recall (DenseNet 0.916, ESPNet 0.866, ResNet 0.807, ShuffleNet 0.753), while ResNet and ESPNet have much faster inference times (ResNet 6 s, ESPNet 8 s, DenseNet 31 s). ResNet is faster than ESPNet, but ESPNet has a higher F-Score and Recall than ResNet, making it a good compromise solution. CONCLUSION: We studied several state-of-the-art CNNs for detecting mitotic figures in whole slide biopsy images. We evaluated two CNNs on a melanoma cancer dataset and then compared four CNNs on a public breast cancer data set, using the same methodology on both. Our methodology and architecture for mitosis finding in both melanoma and breast cancer whole slide images has been thoroughly tested and is likely to be useful for finding mitoses in any whole slide biopsy images.

Candidate enhancers can be identified on the basis of chromatin modifications, the binding of chromatin modifiers and transcription factors and cofactors, or chromatin accessibility. However, validating such candidates as bona fide enhancers requires functional characterization, typically achieved through reporter assays that test whether a sequence can increase expression of a transcriptional reporter via a minimal promoter. A longstanding concern is that reporter assays are mainly implemented on episomes, which are thought to lack physiological chromatin. However, the magnitude and determinants of differences in cis-regulation for regulatory sequences residing in episomes versus chromosomes remain almost completely unknown. To address this systematically, we developed and applied a novel lentivirus-based massively parallel reporter assay (lentiMPRA) to directly compare the functional activities of 2236 candidate liver enhancers in an episomal versus a chromosomally integrated context. We find that the activities of chromosomally integrated sequences are substantially different from the activities of the identical sequences assayed on episomes, and furthermore are correlated with different subsets of ENCODE annotations. The results of chromosomally based reporter assays are also more reproducible and more strongly predictable by both ENCODE annotations and sequence-based models. With a linear model that combines chromatin annotations and sequence information, we achieve a Pearson's R² of 0.362 for predicting the results of chromosomally integrated reporter assays. This level of prediction is better than with either chromatin annotations or sequence information alone and also outperforms predictive models of episomal assays. Our results have broad implications for how cis-regulatory elements are identified, prioritized and functionally validated.

Aim

We sought to discover endogenous urinary biomarkers of human CYP2D6 activity.

Patients & methods

Healthy pediatric subjects (n = 189) were phenotyped using dextromethorphan and randomized for candidate biomarker selection and validation. Global urinary metabolomics was performed using liquid chromatography quadrupole time-of-flight mass spectrometry. Candidate biomarkers were tested in adults receiving fluoxetine, a CYP2D6 inhibitor.

Results

A biomarker, M1 (m/z 444.3102) was correlated with CYP2D6 activity in both the pediatric training and validation sets. Poor metabolizers had undetectable levels of M1, whereas it was present in subjects with other phenotypes. In adult subjects, a 9.56-fold decrease in M1 abundance was observed during CYP2D6 inhibition.

Conclusion

Identification and validation of M1 may provide a noninvasive means of CYP2D6 phenotyping.

The functional consequences of genetic variation in mammalian regulatory elements are poorly understood. We report the in vivo dissection of three mammalian enhancers at single-nucleotide resolution through a massively parallel reporter assay. For each enhancer, we synthesized a library of >100,000 mutant haplotypes with 2-3% divergence from the wild-type sequence. Each haplotype was linked to a unique sequence tag embedded within a transcriptional cassette. We introduced each enhancer library into mouse liver and measured the relative activities of individual haplotypes en masse by sequencing the transcribed tags. Linear regression analysis yielded highly reproducible estimates of the effect of every possible single-nucleotide change on enhancer activity. The functional consequence of most mutations was modest, with ∼22% affecting activity by >1.2-fold and ∼3% by >2-fold. Several, but not all, positions with higher effects showed evidence for purifying selection, or co-localized with known liver-associated transcription factor binding sites, demonstrating the value of empirical high-resolution functional analysis.