SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.01753 · 2022

Self-focusing virtual screening with active design space pruning

arXiv · PDF · Open in the Atlas

Code that ran

12 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 5 failures of our instrument; 0 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/2205.01753.json · badge: /api/ran/2205.01753.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
coleygroup/molpal canonical (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
coleygroup/dsp canonical (classifier label)
classifier basis: classifier_2025
5 ran · 0 no recorded run
FunctionStatusWhere it lives
append_arrays Ran
ran; our contract draft was wrong, not the code
coleygroup/dsp/scripts/collate.py
code served (permissive licence) · get_code("568360a66fa1f1d9")
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 tuple. 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
import tempfile
import os

# Create a temporary npz file with the required keys
temp_dir = tempfile.mkdtemp()
npz_filepath = os.path.join(temp_dir, "test.npz")

# Create sample arrays with distinct shapes for probing
X_data = np.random.rand(2, 4).astype(np.float32)
Y_data = np.random.rand(2, 3).astype(np.float32)
H_data = np.random.rand(2, 5).astype(np.float32)

# Save as npz with the expected keys
np.savez(npz_filepath, X=X_data, Y=Y_data, H=H_data)

# Initialize the list arguments
X_list = []
Y_list = []
Z_list = []

ARGS = {
    "npz_filepath": npz_filepath,
    "X": X_list,
    "Y": Y_list,
    "Z": Z_list,
}
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 tuple — a tuple output has no array shape of its own; none is 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
batches Ran
ran on a synthesized input
coleygroup/molpal/molpal/utils.py
code served (permissive licence) · get_code("42358142a588e362")
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 callable_iterator. 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 torch
import numpy as np
from itertools import islice
from typing import Iterable, Iterator, List

torch.manual_seed(0)

# The function batches() takes an iterable and a size parameter
# It creates batches from the iterable of the given size
# The iterable can be any sequence - let's use a list of diverse items

# Create a simple iterable - a list of numbers
test_iterable = list(range(10))

# Create a batch size parameter
batch_size = 3

ARGS = {
    "it": test_iterable,
    "size": batch_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type callable_iterator — a callable_iterator output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a callable_iterator 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
build_name_dict Ran
ran on a synthesized input
coleygroup/molpal/molpal/cli/extract.py
code served (permissive licence) · get_code("beeecd27d428b9ca")
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 defaultdict. 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 tempfile
import csv
from pathlib import Path
from collections import defaultdict

# Create a temporary directory and CSV file for testing
temp_dir = tempfile.mkdtemp()
parent_dir = Path(temp_dir)

# Create a sample extended.csv file
csv_path = parent_dir / 'extended.csv'
sample_data = [
    ['smi', 'name', 'node_id', 'extra_col'],
    ['CCO', 'ethanol', 'node_1', 'data1'],
    ['CC', 'ethane', 'node_1', 'data2'],
    ['C', 'methane', 'node_2', 'data3'],
    ['CCO', 'ethyl_alcohol', 'node_3', 'data4'],
    ['CCC', 'propane', 'node_2', 'data5'],
]

with open(csv_path, 'w', newline='') as f:
    writer = csv.writer(f)
    writer.writerows(sample_data)

# Define the SMILES to filter by
smis = ['CCO', 'CC', 'C']

ARGS = {
    'parent_dir': parent_dir,
    'smis': smis,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type defaultdict — a defaultdict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a defaultdict 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
calc Ran
ran on a synthesized input
coleygroup/molpal/molpal/acquirer/metrics.py
code served (permissive licence) · get_code("8c90c9a5d3cf36b5")
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 ndarray, ndim 1, dtype family float, shape (6,) 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

# Set seed for reproducibility
np.random.seed(42)

# Create small numpy arrays for Y_mean and Y_var
Y_mean = np.array([0.5, 0.7, 0.3, 0.9, 0.2, 0.6])
Y_var = np.array([0.1, 0.15, 0.08, 0.12, 0.05, 0.11])

# Scalar parameters
current_max = 0.8
t = 0.5  # threshold value
beta = 2  # confidence bound parameter
xi = 0.01  # improvement shift parameter
stochastic = False  # thompson sampling parameter
metric = 'ei'  # example metric

ARGS = {
    'metric': metric,
    'Y_mean': Y_mean,
    'Y_var': Y_var,
    'current_max': current_max,
    't': t,
    'beta': beta,
    'xi': xi,
    'stochastic': stochastic
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (6,) — shape (6,) 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
get_metric Ran
ran on a synthesized input
coleygroup/molpal/molpal/acquirer/metrics.py
code served (permissive licence) · get_code("0f7275ce83ed510e")
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 function. 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

ARGS = {
    'metric': 'ei'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
get_needs Ran
ran on a synthesized input
coleygroup/molpal/molpal/acquirer/metrics.py
code served (permissive licence) · get_code("70126cde14b5ecb5")
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 set. 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 torch
import numpy as np
from typing import Set

torch.manual_seed(0)

# The function get_needs takes a single parameter: metric (a string)
# It returns a Set based on the metric name
# Looking at the function body, it uses a dict.get() with a default of set()
# So we just need to provide valid metric strings

ARGS = {
    'metric': 'greedy'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
mve_loss Ran
ran on a synthesized input
coleygroup/molpal/molpal/models/nnmodels.py
code served (permissive licence) · get_code("5f9a63793db4c090")
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 SymbolicTensor. 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
import tensorflow as tf

# Create tiny tensors for the MVE loss function
batch_size = 2
output_dim = 2

# y_true: ground truth labels, shape (batch_size,)
y_true = tf.constant([1.5, 2.3], dtype=tf.float32)

# y_pred: predicted mean and log-variance, shape (batch_size, 2)
# First column is mu (mean), second column is pre-softplus variance
y_pred = tf.constant([
    [1.2, 0.5],
    [2.1, 0.3]
], dtype=tf.float32)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor output has no array shape of its own; none is 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
pad Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
coleygroup/dsp/scripts/collate.py
code served (permissive licence) · get_code("a9e738a1e258c757")
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 ndarray, ndim 2, dtype family float, shape (5, 4) 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

# Create a 2D array with shape (3, 4) - this will be padded
x = np.array([[1.0, 2.0, 3.0, 4.0],
              [5.0, 6.0, 7.0, 8.0],
              [9.0, 10.0, 11.0, 12.0]])

# Define the target length for padding
length = 5

ARGS = {
    'x': x,
    'length': length
}
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 ndarray, ndim 2, dtype family float, shape (5, 4) — shape (5, 4) 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
plot_acquired_points Ran
ran; our contract draft was wrong, not the code
coleygroup/dsp/scripts/figures.py
code served (permissive licence) · get_code("cd882dcbb6797d2f")
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 PathCollection. 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 torch
import numpy as np
import matplotlib.pyplot as plt

torch.manual_seed(0)

# Create a mock matplotlib axis
fig, ax = plt.subplots()

# X: should be indexable with [:, 0] and [:, 1], so a 2D array/tensor with at least 2 columns
X = np.array([[0.1, 0.2], [0.3, 0.4], [0.5, 0.6]])

# Y: 1D array of values to color the points
Y = np.array([1.0, 2.5, 3.0])

# zmin and zmax: scalar values for the color scale
zmin = 0.5
zmax = 3.5

ARGS = {
    'ax': ax,
    'X': X,
    'Y': Y,
    'zmin': zmin,
    'zmax': zmax
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PathCollection — a PathCollection output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PathCollection 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
plot_surface_discrete Ran
ran; our contract draft was wrong, not the code
coleygroup/dsp/scripts/figures.py
code served (permissive licence) · get_code("effa38928fa09a38")
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 PathCollection. 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
import matplotlib.pyplot as plt
from matplotlib.collections import PathCollection

# Create a figure and axis for plotting
fig, ax = plt.subplots()

# Create sample data for the function
n_points = 10
d = 2

# choices: n x d array of design space points
choices = np.random.rand(n_points, d) * 10

# y: vector of objective values parallel to choices
y = np.random.rand(n_points)

# zmin and zmax: bounds for color scaling
zmin = float(np.min(y))
zmax = float(np.max(y))

# hits: array of selected/hit points in design space
n_hits = 3
hits = np.random.rand(n_hits, d) * 10

# title: optional string for the plot
title = "Test Surface Plot"

ARGS = {
    "ax": ax,
    "choices": choices,
    "y": y,
    "zmin": zmin,
    "zmax": zmax,
    "hits": hits,
    "title": title
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PathCollection — a PathCollection output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PathCollection 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
prune Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
coleygroup/dsp/scripts/figures.py
code served (permissive licence) · get_code("b7cd86bfe4f28fe6")
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 ndarray, ndim 1, dtype family float, shape (5,) 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

# Construct valid arguments for the prune function
choices = np.array([10, 20, 30, 40, 50])
H = np.array([1, 5, 8, 3, 12])
i = 7

ARGS = {
    "choices": choices,
    "H": H,
    "i": i
}
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 ndarray, ndim 1, dtype family float, shape (5,) — shape (5,) 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
read_top_k Ran
ran on a synthesized input
coleygroup/molpal/molpal/cli/extract.py
code served (permissive licence) · get_code("7f37d5583055f1c5")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 tempfile
import csv
from pathlib import Path

# Create a temporary directory structure with the required CSV file
temp_dir = tempfile.TemporaryDirectory()
parent_dir = Path(temp_dir.name)
data_dir = parent_dir / 'data'
data_dir.mkdir(parents=True, exist_ok=True)

# Create the CSV file with sample data
csv_path = data_dir / 'all_explored_final.csv'
with open(csv_path, 'w', newline='') as f:
    writer = csv.writer(f)
    # Write header
    writer.writerow(['SMILES', 'Score'])
    # Write sample data rows
    writer.writerow(['CCO', '0.85'])
    writer.writerow(['CC(C)O', '0.92'])
    writer.writerow(['c1ccccc1', '0.78'])
    writer.writerow(['CC(=O)O', '0.65'])
    writer.writerow(['C1CCCCC1', '0.88'])

ARGS = {
    'parent_dir': parent_dir,
    'k': 3
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 12 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.

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

High-throughput virtual screening is an indispensable technique utilized in the discovery of small molecules. In cases where the library of molecules is exceedingly large, the cost of an exhaustive virtual screen may be prohibitive. Model-guided optimization has been employed to lower these costs through dramatic increases in sample efficiency compared to random selection. However, these techniques introduce new costs to the workflow through the surrogate model training and inference steps. In this study, we propose an extension to the framework of model-guided optimization that mitigates inferences costs using a technique we refer to as design space pruning (DSP), which irreversibly removes poor-performing candidates from consideration. We study the application of DSP to a variety of optimization tasks and observe significant reductions in overhead costs while exhibiting similar performance to the baseline optimization. DSP represents an attractive extension of model-guided optimization that can limit overhead costs in optimization settings where these costs are non-negligible relative to objective costs, such as docking.

For agents

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

get_harvested_code_for_paper("2205.01753")
get_code_for_paper("2205.01753")
have("2205.01753")

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

curl https://syntology.ai/api/ran/2205.01753.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.01753.svg)](https://syntology.ai/paper/2205.01753)

Connect an agent — have() is free.