Boosting Local Causal Discovery in High-Dimensional Expression Data

Versteeg, Philip; Mooij, Joris M.

doi:10.1109/bibm47256.2019.8983232

Cited by 7 publications

(10 citation statements)

References 16 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This serves as a benchmark, mostly of interest in the simulations. and Figure 4 of Versteeg and Mooij [2019]. The significant enrichment of causal effects in the top 5 or 10 predictions made by ICP and CD was seen in these works as evidence that these methods are identifying causal relations.…”

Section: Kemmerenmentioning

confidence: 73%

“…Kemmeren has been used to evaluate the predictive performance of several causal models including ICP, the CD and and LCD , Rothenhäusler et al, 2019, Versteeg and Mooij, 2019. The performance of these models has been compared to purely association based methods such as Lasso regression.…”

Section: Kemmerenmentioning

confidence: 99%

“…Following , we say that gene i has a significant causal effect on gene j if in the sample where gene i is knocked out, the expression of gene j is outside of the range of gene j measured in the n 1 observational samples. See Versteeg and Mooij [2019] and for other possible definitions of causal effect.…”

Section: Kemmerenmentioning

confidence: 99%

“…This strategy was used to compare performance of several recently proposed causal models on a yeast knockout data set produced in Kemmeren et al [2014] (hereafter Kemmeren). The causal models assessed include Invariant Causal Prediction (ICP) and the Causal Dantzig (CD) which fit regression models which are invariant across data collection environments [Rothenhäusler et al, 2019 and Local Causal Discovery (LCD) which tests for dependence among gene expression and context variables [Versteeg and Mooij, 2019]. These models were compared to standard (non-causal) models such as Lasso regression.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Sample Selection Bias in Evaluation of Prediction Performance of Causal Models

Long,

2021

Preprint

View full text Add to dashboard Cite

Causal models are notoriously difficult to validate because they make untestable assumptions regarding confounding. New scientific experiments offer the possibility of evaluating causal models using prediction performance. Prediction performance measures are typically robust to violations in causal assumptions. However prediction performance does depend on the selection of training and test sets. In particular biased training sets can lead to optimistic assessments of model performance. In this work, we revisit the prediction performance of several recently proposed causal models tested on a genetic perturbation data set of Kemmeren [Kemmeren et al., 2014]. We find that sample selection bias is likely a key driver of model performance. We propose using a less-biased evaluation set for assessing prediction performance on Kemmeren and compare models on this new set. In this setting, the causal model tested have similar performance to standard association based estimators such as Lasso. Finally we compare the performance of causal estimators in simulation studies which reproduce the Kemmeren structure of genetic knockout experiments but without any sample selection bias. These results provide an improved understanding of the performance of several causal models and offer guidance on how future studies should use Kemmeren.

show abstract

Section: Kemmerenmentioning

confidence: 73%

Section: Kemmerenmentioning

confidence: 99%

Section: Kemmerenmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Sample Selection Bias in Evaluation of Prediction Performance of Causal Models

Long,

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In Kemmeren et al (2014), a microarray dataset with gene expressions is introduced, which has been used for the real-world evaluation of causal discovery methods such as LCD (Versteeg and Mooij, 2019). shows the successful application of the ICP algorithm on this dataset.…”

Section: Related Workmentioning

confidence: 99%

Local Constraint-Based Causal Discovery under Selection Bias

Versteeg¹,

Zhang²,

Mooij³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

We consider the problem of discovering causal relations from independence constraints selection bias in addition to confounding is present. While the seminal FCI algorithm is sound and complete in this setup, no criterion for the causal interpretation of its output under selection bias is presently known. We focus instead on local patterns of independence relations, where we find no sound method for only three variable that can include background knowledge. Y-Structure patterns (Mani et al., 2006;Mooij and Cremers, 2015) are shown to be sound in predicting causal relations from data under selection bias, where cycles may be present. We introduce a finite-sample scoring rule for Y-Structures that is shown to successfully predict causal relations in simulation experiments that include selection mechanisms. On real-world microarray data, we show that a Y-Structure variant performs well across different datasets, potentially circumventing spurious correlations due to selection bias.

show abstract

Sample selection bias in evaluation of prediction performance of causal models

Long

2021

Statistical Analysis

View full text Add to dashboard Cite

Causal models are notoriously difficult to validate because they make untestable assumptions regarding confounding. New scientific experiments offer the possibility of evaluating causal models using prediction performance. Prediction performance measures are typically robust to violations in causal assumptions. However, prediction performance does depend on the selection of training and test sets. Biased training sets can lead to optimistic assessments of model performance. In this work, we revisit the prediction performance of several recently proposed causal models tested on a genetic perturbation data set of Kemmeren. We find that sample selection bias is likely a key driver of model performance. We propose using a less‐biased evaluation set for assessing prediction performance and compare models on this new set. In this setting, the causal models have similar or worse performance compared to standard association‐based estimators such as Lasso. Finally, we compare the performance of causal estimators in simulation studies that reproduce the Kemmeren structure of genetic knockout experiments but without any sample selection bias. These results provide an improved understanding of the performance of several causal models and offer guidance on how future studies should use Kemmeren.

show abstract

Boosting Local Causal Discovery in High-Dimensional Expression Data

Cited by 7 publications

References 16 publications

Sample Selection Bias in Evaluation of Prediction Performance of Causal Models

Sample Selection Bias in Evaluation of Prediction Performance of Causal Models

Local Constraint-Based Causal Discovery under Selection Bias

Sample selection bias in evaluation of prediction performance of causal models

Contact Info

Product

Resources

About