Causal Sufficiency Research Articles

Time Delayed Causal Gene Regulatory Network Inference with Hidden Common Causes.

Inferring the gene regulatory network (GRN) is crucial to understanding the working of the cell. Many computational methods attempt to infer the GRN from time series expression data, instead of through expensive and time-consuming experiments. However, existing methods make the convenient but unrealistic assumption of causal sufficiency, i.e. all the relevant factors in the causal network have been observed and there are no unobserved common cause. In principle, in the real world, it is impossible to be certain that all relevant factors or common causes have been observed, because some factors may not have been conceived of, and therefore are impossible to measure. In view of this, we have developed a novel algorithm named HCC-CLINDE to infer an GRN from time series data allowing the presence of hidden common cause(s). We assume there is a sparse causal graph (possibly with cycles) of interest, where the variables are continuous and each causal link has a delay (possibly more than one time step). A small but unknown number of variables are not observed. Each unobserved variable has only observed variables as children and parents, with at least two children, and the children are not linked to each other. Since it is difficult to obtain very long time series, our algorithm is also capable of utilizing multiple short time series, which is more realistic. To our knowledge, our algorithm is far less restrictive than previous works. We have performed extensive experiments using synthetic data on GRNs of size up to 100, with up to 10 hidden nodes. The results show that our algorithm can adequately recover the true causal GRN and is robust to slight deviation from Gaussian distribution in the error terms. We have also demonstrated the potential of our algorithm on small YEASTRACT subnetworks using limited real data.

Surrogate Endpoint Evaluation: Principal Stratification Criteria and the Prentice Definition.

A common problem of interest within a randomized clinical trial is the evaluation of an inexpensive response endpoint as a valid surrogate endpoint for a clinical endpoint, where a chief purpose of a valid surrogate is to provide a way to make correct inferences on clinical treatment effects in future studies without needing to collect the clinical endpoint data. Within the principal stratification framework for addressing this problem based on data from a single randomized clinical efficacy trial, a variety of definitions and criteria for a good surrogate endpoint have been proposed, all based on or closely related to the "principal effects" or "causal effect predictiveness (CEP)" surface. We discuss CEP-based criteria for a useful surrogate endpoint, including (1) the meaning and relative importance of proposed criteria including average causal necessity (ACN), average causal sufficiency (ACS), and large clinical effect modification; (2) the relationship between these criteria and the Prentice definition of a valid surrogate endpoint; and (3) the relationship between these criteria and the consistency criterion (i.e., assurance against the "surrogate paradox"). This includes the result that ACN plus a strong version of ACS generally do not imply the Prentice definition nor the consistency criterion, but they do have these implications in special cases. Moreover, the converse does not hold except in a special case with a binary candidate surrogate. The results highlight that assumptions about the treatment effect on the clinical endpoint before the candidate surrogate is measured are influential for the ability to draw conclusions about the Prentice definition or consistency. In addition, we emphasize that in some scenarios that occur commonly in practice, the principal strata sub-populations for inference are identifiable from the observable data, in which cases the principal stratification framework has relatively high utility for the purpose of effect modification analysis, and is closely connected to the treatment marker selection problem. The results are illustrated with application to a vaccine efficacy trial, where ACN and ACS for an antibody marker are found to be consistent with the data and hence support the Prentice definition and consistency.

Targeted ablation of cholinergic interneurons in the dorsolateral striatum produces behavioral manifestations of Tourette syndrome

Gilles de la Tourette syndrome (TS) is characterized by tics, which are transiently worsened by stress, acute administration of dopaminergic drugs, and by subtle deficits in motor coordination and sensorimotor gating. It represents the most severe end of a spectrum of tic disorders that, in aggregate, affect ∼ 5% of the population. Available treatments are frequently inadequate, and the pathophysiology is poorly understood. Postmortem studies have revealed a reduction in specific striatal interneurons, including the large cholinergic interneurons, in severe disease. We tested the hypothesis that this deficit is sufficient to produce aspects of the phenomenology of TS, using a strategy for targeted, specific cell ablation in mice. We achieved ∼ 50% ablation of the cholinergic interneurons of the striatum, recapitulating the deficit observed in patients postmortem, without any effect on GABAergic markers or on parvalbumin-expressing fast-spiking interneurons. Interneuron ablation in the dorsolateral striatum (DLS), corresponding roughly to the human putamen, led to tic-like stereotypies after either acute stress or d-amphetamine challenge; ablation in the dorsomedial striatum, in contrast, did not. DLS interneuron ablation also led to a deficit in coordination on the rotorod, but not to any abnormalities in prepulse inhibition, a measure of sensorimotor gating. These results support the causal sufficiency of cholinergic interneuron deficits in the DLS to produce some, but not all, of the characteristic symptoms of TS.

Discovering and orienting the edges connected to a target variable in a DAG via a sequential local learning approach

Given a target variable and observational data, we propose a sequential learning approach for discovering direct cause and effect variables of the target under the causal network framework. In the approach, we start from the target, sequentially find Markov blankets of variables and learn local structures over Markov blankets until we determine the causes and effects of the target variable. Without constructing a whole network over all variables, we find only a local structure around the target. The main advantage of the proposed sequential approach is that it can greatly reduce CPU time compared with whole network learning approaches for finding the causes and effects of a given target node in a large network. The proposed approach can be applied to predict the effects of external interventions. Theoretically we show the correctness of the approach under the assumptions of faithfulness, causal sufficiency and that independencies are correctly checked. These theoretical results can also be used for learning directed acyclic graphs with latent variables.

Safety vs. efficacy assessment of pharmaceuticals: Epistemological rationales and methods

In their comparative analysis of Randomised Clinical Trials and observational studies, Papanikoloau et al. (2006) assert that “it may be unfair to invoke bias and confounding to discredit observational studies as a source of evidence on harms”. There are two kinds of answers to the question why this is so. One is based on metaphysical assumptions, such as the problem of causal sufficiency, modularity and other statistical assumptions. The other is epistemological and relates to foundational issues and how they determine the constraints we put on evidence. I will address here the latter dimension and present recent proposals to amend evidence hierarchies for the purpose of safety assessment of pharmaceuticals; I then relate these suggestions to a case study: the recent debate on the causal association between paracetamol and asthma. The upshot of this analysis is that different epistemologies impose different constraints on the methods we adopt to collect and evaluate evidence; thus they grant “lower level” evidence on distinct grounds and at different conditions. Appreciating this state of affairs illuminates the debate on the epistemic asymmetry concerning benefits and harms and sets the basis for a foundational, as opposed to heuristic, justification of safety assessment based on heterogeneous evidence.

Exclusion, still not tracted

Karen Bennett has recently articulated and defended a “compatibilist” solution to the causal exclusion problem. Bennett’s solution works by rejecting the exclusion principle on the grounds that even though physical realizers are distinct from the mental states or properties that they realize, they necessarily co-occur such that they fail to satisfy standard accounts of causal over-determination. This is the case, Bennett argues, because the causal background conditions for core realizers being sufficient causes of their effects are identical to the “surround” conditions with which the core realizers are metaphysically sufficient the states or properties that they realize. Here we demonstrate that the background conditions for the causal sufficiency of core realizers for their effects are not identical to the core realizer’s surrounds, nor do backgrounds necessitate such surround conditions. If compatibilist solutions to exclusion can be defended, a different argument will be needed.

Causality in Linear Nongaussian Acyclic Models in the Presence of Latent Gaussian Confounders

LiNGAM has been successfully applied to some real-world causal discovery problems. Nevertheless, causal sufficiency is assumed; that is, there is no latent confounder of the observations, which may be unrealistic for real-world problems. Taking into the consideration latent confounders will improve the reliability and accuracy of estimations of the real causal structures. In this letter, we investigate a model called linear nongaussian acyclic models in the presence of latent gaussian confounders (LiNGAM-GC) which can be seen as a specific case of lvLiNGAM. This model includes the latent confounders, which are assumed to be independent gaussian distributed and statistically independent of the disturbances. To tackle the causal discovery problem of this model, first we propose a pairwise cumulant-based measure of causal directions for cause-effect pairs. We prove that in spite of the presence of latent gaussian confounders, the causal direction of the observed cause-effect pair can be identified under the mild condition that the disturbances are simultaneously supergaussian or subgaussian. We propose a simple and efficient method to detect the violation of this condition. We extend our work to multivariate causal network discovery problems. Specifically we propose algorithms to estimate the causal network structure, including causal ordering and causal strengths, using an iterative root finding-removing scheme based on pairwise measure. To address the redundant edge problem due to the finite sample size effect, we develop an efficient bootstrapping-based pruning algorithm. Experiments on synthetic data and real-world data have been conducted to show the applicability of our model and the effectiveness of our proposed algorithms.

Discovering causes and effects of a given node in Bayesian networks

Causal relationships among variables can be depicted by a causal network of these variables. We propose a local structure learning approach for discovering the direct causes and the direct effects of a given target variable. In the approach, we first find the variable set of parents, children, and maybe some descendants (PCD) of the target variable, but generally we cannot distinguish the parents from the children in the PCD of the target variable. Next, to distinguish the causes from the effects of the target variable, we find the PCD of each variable in the PCD of the target variable, and we repeat the process of finding PCDs along the paths starting from the target variable. Without constructing a whole network over all variables, we find only a local structure around the target variable. Theoretically, we show the correctness of the proposed approach under the assumptions of faithfulness, causal sufficiency, and that conditional independencies are correctly checked.

Conservative independence-based causal structure learning in absence of adjacency faithfulness

This paper presents an extension to the Conservative PC algorithm which is able to detect violations of adjacency faithfulness under causal sufficiency and triangle faithfulness. Violations can be characterized by pseudo-independent relations and equivalent edges, both generating a pattern of conditional independencies that cannot be modeled faithfully. Both cases lead to uncertainty about specific parts of the skeleton of the causal graph. These ambiguities are modeled by an f-pattern. We prove that our Adjacency Conservative PC algorithm is able to correctly learn the f-pattern. We argue that the solution also applies for the finite sample case if we accept that only strong edges can be identified. Experiments based on simulations and the ALARM benchmark model show that the rate of false edge removals is significantly reduced, at the expense of uncertainty on the skeleton and a higher sensitivity for accidental correlations.

Physicalism, Emergence and Downward Causation

The development of a defensible and fecund notion of emergence has been dogged by a number of threshold issues neatly highlighted in a recent paper by Jaegwon Kim. We argue that physicalist assumptions confuse and vitiate the whole project. In particular, his contention that emergence entails supervenience is contradicted by his own argument that the ‘microstructure’ of an object belongs to the whole object, not to its constituents. And his argument against the possibility of downward causation is question-begging and makes false assumptions about causal sufficiency. We argue, on the contrary, for a rejection of the deeply entrenched assumption, shared by physicalists and Cartesians alike, that what basically exists are things (entities, substances). Our best physics tells us that there are no basic particulars, only fields in process. We need an ontology which gives priority to organization, which is inherently relational. Reflection upon the fact that all biological creatures are far-from-equilibrium systems, whose very persistence depend upon their interactions with their environment, reveals incoherence in the notion of an ‘emergence base’.

Causal Exclusion and the Preservation of Causal Sufficiency

The causal exclusion argument against the causal efficacy of non-reducible mental properties relies on a causal closure principle for the physical domain, and on a causal exclusion principle ruling out systematic causal overdetermination. The notion of causal sufficiency is typically used in formulating both these principles. I argue that the causal exclusion principle has to be substantially revised in light of counterexamples where causal sufficiency is preserved among closely related cause candidates. This crucially affects the causal exclusion argument, and I specify which among several versions of the causal closure principle are compatible with the causal efficacy of mental properties.

PPARα siRNA–Treated Expression Profiles Uncover the Causal Sufficiency Network for Compound-Induced Liver Hypertrophy

Uncovering pathways underlying drug-induced toxicity is a fundamental objective in the field of toxicogenomics. Developing mechanism-based toxicity biomarkers requires the identification of such novel pathways and the order of their sufficiency in causing a phenotypic response. Genome-wide RNA interference (RNAi) phenotypic screening has emerged as an effective tool in unveiling the genes essential for specific cellular functions and biological activities. However, eliciting the relative contribution of and sufficiency relationships among the genes identified remains challenging. In the rodent, the most widely used animal model in preclinical studies, it is unrealistic to exhaustively examine all potential interactions by RNAi screening. Application of existing computational approaches to infer regulatory networks with biological outcomes in the rodent is limited by the requirements for a large number of targeted permutations. Therefore, we developed a two-step relay method that requires only one targeted perturbation for genome-wide de novo pathway discovery. Using expression profiles in response to small interfering RNAs (siRNAs) against the gene for peroxisome proliferator-activated receptor α (Ppara), our method unveiled the potential causal sufficiency order network for liver hypertrophy in the rodent. The validity of the inferred 16 causal transcripts or 15 known genes for PPARα-induced liver hypertrophy is supported by their ability to predict non-PPARα–induced liver hypertrophy with 84% sensitivity and 76% specificity. Simulation shows that the probability of achieving such predictive accuracy without the inferred causal relationship is exceedingly small (p < 0.005). Five of the most sufficient causal genes have been previously disrupted in mouse models; the resulting phenotypic changes in the liver support the inferred causal roles in liver hypertrophy. Our results demonstrate the feasibility of defining pathways mediating drug-induced toxicity from siRNA-treated expression profiles. When combined with phenotypic evaluation, our approach should help to unleash the full potential of siRNAs in systematically unveiling the molecular mechanism of biological events.

Free will, the self, and the brain

The free will problem is defined and three solutions are discussed: no-freedom theory, libertarianism, and compatibilism. Strict determinism is often assumed in arguing for libertarianism or no-freedom theory. It assumes that the history of the universe is fixed, but modern physics admits a certain degree of randomness in the determination of events. However, this is not enough for a compatibilist position-which is favored here-since freedom is not randomness. It is the I that chooses what to do. It is argued that the core of the free will problem is what this I is. A materialist view is favored: The I is an activity of the brain. In addition to absence of external and internal compulsion, freedom involves absence of causal sufficiency of influences acting on the I. A more elaborate compatibilist view is proposed, according to which causal determination is complete when we add events occurring in the I (of which the subject is not conscious). Contrary to what several authors have argued, the onset of the readiness potential before the decision to act is no problem here. The experience of agency is incomplete and fallible, rather than illusory. Some consequences of different views about freedom for the ascription of responsibility are discussed.

PPARα siRNA-Treated Expression Profiles Uncover the Causal Sufficiency Network for Compound-Induced Liver Hypertrophy

PPARα siRNA-Treated Expression Profiles Uncover the Causal Sufficiency Network for Compound-Induced Liver Hypertrophy

Predictive Inference Generation as a Function of Working Memory Capacity and Causal Text Constraints

Two factors that affect the predictive inferential process are examined: the working memory (WM) capacity of the reader and causal text constraints. A naming task was used in Experiment 1 to investigate predictive inferences that were based on texts containing low, moderate, and high causal sufficiency events at 2 time delays (250 and 500 ms). The results indicated that high WM capacity readers make predictive inferences when texts contain high causal sufficiency events, at both time delays, as evidenced by naming time differences between high-low and high-moderate causal sufficiency conditions. There is no evidence that low WM capacity readers make predictive inferences at either time delay. In Experiment 2, reading times for target sentences that either confirmed or contradicted predictive inferences in low, moderate, and high causal sufficiency conditions were collected. The results showed that high WM capacity readers make predictive inferences when causal sufficiency is high and target sentences confirmatory and that both low and high WM capacity readers make predictive inferences based on the comparison between high causal sufficiency texts in the confirmatory versus contradictory conditions. Thus, both experiments demonstrate that generating predictive inferences is a demanding process that requires a certain amount of WM resources and causal text constraints.

Once More into the Bramble Bush: Duty, Causal Contribution, and the Extent of Legal Responsibility

Courts, lawyers, law students, and academics continue to confuse the empirical issue of causal contribution with the distinct normative issues of tortious conduct and legal injury, which precede and frame the causal-contribution inquiry, and the normative issue of the extent of legal responsibility for tortiously caused consequences, which follows the causal-contribution inquiry. In a number of prior articles, I have tried to distinguish and clarify these various issues, which arise not only in tort law, but also in much the same form in criminal law and many other areas of the law. I have focused primarily on distinguishing and clarifying the empirical issue of causal contribution and elaborating a comprehensive test, the test, for resolving this issue. In this paper, which was prepared for the recent Wade Conference on the Third Restatement of Torts: General Principles, I revisit these issues. I focus more than I previously have on the Restatement's unhelpful, opaque, confused, and contradictory treatments of these issues, while also commenting on recent scholarship which fails to properly distinguish these issues. I defend the NESS test of causal contribution against some recent criticisms, propose a practical way of properly presenting the causal-contribution issue to students and jurors, criticize alternative proposed tests (including Jane Stapleton's targeted but-for test), further elaborate the notion of causal sufficiency (rather than mere analytical or empirical sufficiency) that underlies the NESS test, and provide a more detailed explanation of the NESS test's application to the conceptually most difficult types of causation cases, the overdetermined multiple-omission cases.

Causation and the Law: Preemption, Lawful Sufficiency, and Causal Sufficiency

RICHARD FUMERTON [*] KEN KRESS [**] I INTRODUCTION Richard Wright's attempt to distinguish actual causation from proximate causation, and to analyze actual causation--particularly in tort law--has been the most successful and influential work in this area in recent years. [1] Indeed, most scholars think that Wright's analysis of cause as a necessary element in a set of conditions (NESS) sufficient an effect [2] is an improvement upon H.L.A. Hart and Tony Honore's classic discussion. [3] In the decade and a half since Wright put forth his theory of causation, there have been a number of attempts to criticize him, none has undermined the central thrust of his theory. [4] While Wright's view has received deserved praise, his analysis of a NESS requires clarification. Once clarified, however, it is vulnerable to certain fundamental objections. Wright's stated aim is to capture a nonnormative concept of causation. Therefore, this article begins by briefly describing the normative/nonnormative distinction, and how one might invoke this distinction to locate a nonnormative dimension of actual causation (as opposed to proximate causation). After briefly introducing Wright's concept of a NESS an effect (the concept in terms of which he wants to understand actual causation), the article notes ambiguities in the critical concepts of necessity and sufficiency that he deploys. The article distinguishes a number of different modal concepts and suggests the most plausible interpretation of Wright's use of these concepts. With a more precise understanding of Wright's view in hand, we turn to the question of whether his analysis more effectively handles difficult cases of causation--including multiple causes, preemption, omissions and the like--than does the nearly universal but for test deployed by courts as a test of actual causation. While Wright's test has certain advantages over the but for test, the article argues that it faces difficult problems of its own. First, it is not clear that Wright can accommodate indeterministic causation. Nor, contrary to what he claims, can his account adequately explain all cases of overdetermination involving preemption. The most obvious revision of Wright's theory to handle cases of preemption (replacing a concept of lawful sufficiency with one of causal sufficiency--a revision with which Wright has indicated sympathy) risks vicious conceptual circularity. It hardly seems likely that one will gain insight into the nature of causation through an account that relies on an understanding of that special kind of sufficiency involved in causation. II CAUSATION AND NORMATIVITY Legal theorists have tried to distinguish nonnormative issues from those that include a normative element. [5] Obvious candidates nonnormative issues are questions concerning causation. For example, if you are suing me damages in tort, one of the things you must establish your suit to be successful is that some action I took caused the relevant damages. [6] This part of your burden seems straightforwardly factual in a way in which, say, your burden of showing that I acted negligently (or tortiously) may not be. It seems plausible on its face to suppose that negligence is a normative notion. When people act negligently, they do not exercise due care. They do not take precautions they ought to take in the face of various possible harms. The italicized terms are paradigm value terms that indicate our subject has taken a normative turn. Before proceeding, it might be worth elaborating on this distinction between nonnormative and normative questions. Some people describe nonnormative issues as factual and normative issues as nonfactual. Many ethical philosophers resist this way of presenting the contrast. These philosophers argue that normative, ethical judgments are factual judgments. Some of these descriptivists are subjectivists, others would claim that ethical statements make objective assertions. …

Counterfactual thinking and regulatory focus: Implications for action versus inaction and sufficiency versus necessity.

Counterfactual thinking is associated with regulatory focus in a way that explains previous empirical incongruities, such as whether additive counterfactuals (mutations of inactions) occur more or less frequently than subtractive counterfactuals (mutations of actions). In Experiment 1, regulatory focus moderated this pattern, in that additive counterfactuals were activated by promotion failure, whereas subtractive counterfactuals were activated by prevention failure. In Experiment 2, additive counterfactuals evoked a promotion focus and expressed causal sufficiency, whereas subtractive counterfactuals evoked a prevention focus and expressed causal necessity. In Experiment 3, dejection activated additive counterfactuals, whereas agitation activated subtractive counterfactuals. These findings illuminate the interconnections among counterfactual thinking, motivation, and goals.

Counterfactual thinking and regulatory focus: implications for action versus inaction and sufficiency versus necessity.

Counterfactual thinking is associated with regulatory focus in a way that explains previous empirical incongruities, such as whether additive counterfactuals (mutations of inactions) occur more or less frequently than subtractive counterfactuals (mutations of actions). In Experiment 1, regulatory focus moderated this pattern, in that additive counterfactuals were activated by promotion failure, whereas subtractive counterfactuals were activated by prevention failure. In Experiment 2, additive counterfactuals evoked a promotion focus and expressed causal sufficiency, whereas subtractive counterfactuals evoked a prevention focus and expressed causal necessity. In Experiment 3, dejection activated additive counterfactuals, whereas agitation activated subtractive counterfactuals. These findings illuminate the interconnections among counterfactual thinking, motivation, and goals.

Calvin and Bernard On Freedom and Necessity: A Reply to Brümmer

It is argued that Calvin does not veer between two incompatible accounts of grace, freedom and necessity in Institutes II. 2, but presents a consistent position. The consistency is evident once it is seen that Calvin carefully distinguished between necessity and compulsion. For him not all necessitated acts are compelled, but all human acts which are the outcome of efficacious divine grace are necessitated by that grace. Because Calvin is consistent, there is no need to suppose that he has mistaken the causal sufficiency of divine saving grace for its causal importance.