SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.15500 · ICLR · 2024

Gene Regulatory Network Inference in the Presence of Dropouts: a Causal View

Kun Zhang, Ignavier Ng, Haoyue Dai, Peter Spirtes, Petar Stojanov, Gongxu Luo

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 3 violated it and 1 has no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2403.15500.json · badge: /api/ran/2403.15500.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
markdana/scrna-causal-dropout canonical (classifier label)
classifier basis: classifier_2025
4 ran · 1 no recorded run
FunctionStatusWhere it lives
LocalScoreClass Ran
ran on a synthesized input · constructed an object; no result was computed
markdana/scrna-causal-dropout/csl/score/LocalScoreFunctionClass.py
code served (permissive licence) · get_code("d534d601fcb83279")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, LocalScoreClass: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from typing import Any, Callable, Dict, List, Tuple

# Create small sample data
np.random.seed(0)
data = np.random.randn(10, 4)  # 10 samples, 4 variables

# Define the local score function
def local_score_BIC_from_cov(Data: Tuple[np.ndarray, int], i: int, PAi: List[int], parameters=None) -> float:
    """
    Calculate the *negative* local score with BIC for the linear Gaussian continue data case
    from the covariance matrix
    """
    cov, n = Data
    if parameters is None:
        lambda_value = 1
    else:
        lambda_value = parameters['lambda_value']
    if len(PAi) == 0:
        return n * np.log(cov[i, i])
    yX = np.mat(cov[np.ix_([i], PAi)])
    XX = np.mat(cov[np.ix_(PAi, PAi)])
    H = np.log(cov[i, i] - yX * np.linalg.inv(XX) * yX.T)
    return n * H + np.log(n) * len(PAi) * lambda_value

# Define LocalScoreClass
class LocalScoreClass(object):

    def __init__(self, data: Any, local_score_fun: Callable[[Any, int, List[int], Any], float], parameters=None):
        self.data = data
        self.local_score_fun = local_score_fun
        self.parameters = parameters
        self.score_cache = {}
        if self.local_score_fun == local_score_BIC_from_cov:
            self.cov = np.cov(self.data.T)
            self.n = self.data.shape[0]

    def score(self, i: int, PAi: List[int]) -> float:
        if i not in self.score_cache:
            self.score_cache[i] = {}
        hash_key = tuple(sorted(PAi))
        if not self.score_cache[i].__contains__(hash_key):
            if self.local_score_fun == local_score_BIC_from_cov:
                self.score_cache[i][hash_key] = self.local_score_fun((self.cov, self.n), i, PAi, self.parameters)
            else:
                self.score_cache[i][hash_key] = self.local_score_fun(self.data, i, PAi, self.parameters)
        return self.score_cache[i][hash_key]

# Set up arguments for LocalScoreClass
ARGS = {
    'data': data,
    'local_score_fun': local_score_BIC_from_cov,
    'parameters': {'lambda_value': 1.0}
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type LocalScoreClass — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a LocalScoreClass output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
bic_score_change Ran
ran, violated the contract we drafted
markdana/scrna-causal-dropout/csl/search/ScoreBased/GES.py
code served (permissive licence) · get_code("39e0fe1787fe6f6c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned float64, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function bic_score_change takes two parameters:
# - rho: a numeric value (correlation coefficient, typically between -1 and 1)
# - N: a numeric value (sample size, typically a positive integer)

# Looking at the function body:
# - rho is used in: 1 - rho ** 2 (arithmetic operations)
# - N is used in: np.log(1 - rho ** 2) multiplied by N, and np.log(N) (arithmetic operations)

# Both parameters are numeric scalars used in mathematical operations

ARGS = {
    "rho": 0.5,  # correlation coefficient between -1 and 1
    "N": 100,    # sample size (positive integer)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
fisherz_rho Ran
ran, violated the contract we drafted
markdana/scrna-causal-dropout/csl/search/ScoreBased/GES.py
code served (permissive licence) · get_code("79be0214f2518978")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned float64, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Analyze the function signature and body
# fisherz_rho(stat, N)
# - stat: used in arithmetic operations (2 * stat / np.sqrt(N - 3))
# - N: used in arithmetic operations (N - 3)
# Both appear to be numeric values

# Create valid arguments
stat = 0.5  # A scalar float value
N = 10      # A scalar integer value (must be > 3 for the sqrt to be valid)

ARGS = {
    "stat": stat,
    "N": N
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
fisherz_stat Ran
ran, violated the contract we drafted
markdana/scrna-causal-dropout/csl/search/ScoreBased/GES.py
code served (permissive licence) · get_code("0cb4f1fc8a3a0791")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned float64, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function fisherz_stat takes two parameters:
# - rho: a correlation coefficient (float between -1 and 1)
# - N: sample size (integer)

# Create valid arguments
rho = 0.65  # correlation coefficient between -1 and 1
N = 50      # sample size

ARGS = {
    "rho": rho,
    "N": N
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
local_score_BIC_from_cov Not yet run
unverified (harvested, no recorded run)
markdana/scrna-causal-dropout/csl/score/LocalScoreFunctionClass.py
code served (permissive licence) · get_code("51aba01784afcab9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
local_score_BIC_from_cov
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2402.13241 (2024-02)
Federated Causal Discovery from Heterogeneous Data

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Gene regulatory network inference (GRNI) is a challenging problem, particularly owing to the presence of zeros in single-cell RNA sequencing data: some are biological zeros representing no gene expression, while some others are technical zeros arising from the sequencing procedure (aka dropouts), which may bias GRNI by distorting the joint distribution of the measured gene expressions. Existing approaches typically handle dropout error via imputation, which may introduce spurious relations as the true joint distribution is generally unidentifiable. To tackle this issue, we introduce a causal graphical model to characterize the dropout mechanism, namely, Causal Dropout Model. We provide a simple yet effective theoretical result: interestingly, the conditional independence (CI) relations in the data with dropouts, after deleting the samples with zero values (regardless if technical or not) for the conditioned variables, are asymptotically identical to the CI relations in the original data without dropouts. This particular test-wise deletion procedure, in which we perform CI tests on the samples without zeros for the conditioned variables, can be seamlessly integrated with existing structure learning approaches including constraint-based and greedy score-based methods, thus giving rise to a principled framework for GRNI in the presence of dropouts. We further show that the causal dropout model can be validated from data, and many existing statistical models to handle dropouts fit into our model as specific parametric instances. Empirical evaluation on synthetic, curated, and real-world experimental transcriptomic data comprehensively demonstrate the efficacy of our method.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2403.15500")
get_code_for_paper("2403.15500")
have("2403.15500")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2403.15500.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2403.15500.svg)](https://syntology.ai/paper/2403.15500)

Connect an agent — have() is free.