SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.11248 · 2020

Accelerating Antimicrobial Discovery with Controllable Deep Generative Models and Molecular Dynamics

arXiv · PDF · Open in the Atlas

Code that ran

11 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 11 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2005.11248.json · badge: /api/ran/2005.11248.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
IBM/amp-design community (classifier label)
classifier basis not recorded on the link
11 ran · 1 no recorded run
FunctionStatusWhere it lives
build_classifier Ran
ran on a synthesized input
IBM/amp-design/models/classifier.py
code served (permissive licence) · get_code("a09ca7dd6c752c32")
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 CNNClassifier. 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 torch.nn as nn

torch.manual_seed(0)

# Set up parameters for build_classifier
classifier_type = 'cnn'
emb_dim = 4
min_filter_width = 2
max_filter_width = 5
num_filters = 3
dropout = 0.5

ARGS = {
    'classifier_type': classifier_type,
    'emb_dim': emb_dim,
    'min_filter_width': min_filter_width,
    'max_filter_width': max_filter_width,
    'num_filters': num_filters,
    'dropout': dropout
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CNNClassifier — a CNNClassifier output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CNNClassifier 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_encoder Ran
ran on a synthesized input
IBM/amp-design/models/encoder.py
code served (permissive licence) · get_code("ea5e7e863df2c31e")
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 GRUEncoder. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'enc_type': 'gru',
    'emb_dim': 4,
    'h_dim': 8,
    'z_dim': 3,
    'biGRU': True,
    'layers': 2,
    'p_dropout': 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type GRUEncoder — a GRUEncoder output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a GRUEncoder 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
check_mask_eos Ran
ran on a synthesized input
IBM/amp-design/models/mutils.py
code served (permissive licence) · get_code("9020274a07f64952")
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 int, ndim 0, dtype family int, 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 torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.EOS_IDX = 2
        self.PAD_IDX = 0

model = SimpleModel()

# Create a sentence tensor with shape (1, 5)
# Contains some tokens, an EOS token, and padding after EOS
sentence = torch.tensor([[1, 1, 2, 0, 0]], dtype=torch.long)

ARGS = {
    "sentence": sentence,
    "model": model
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, 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
encode_sequence Ran
ran on a synthesized input
IBM/amp-design/api.py
code served (permissive licence) · get_code("67a198e2164bc6ca")
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 Tensor, ndim 2, dtype family float, shape (1, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Stand-in vocab class with to_ix method
class Vocab:
    def to_ix(self, sequence):
        # Return a tensor of token indices
        # sequence is a string, convert to dummy indices
        indices = [ord(c) % 10 for c in sequence]
        return torch.tensor(indices, dtype=torch.long)

# Stand-in model with forward_encoder and sample_z methods
class VAEModel(nn.Module):
    def __init__(self, latent_dim=4):
        super().__init__()
        self.latent_dim = latent_dim
        self.encoder = nn.Linear(10, latent_dim * 2)
    
    def forward_encoder(self, enc_inputs):
        # enc_inputs is a 1D tensor of indices
        # Embed and encode to get mu and logvar
        batch_size = 1
        # Simple embedding: one-hot encode the indices
        embedded = torch.zeros(10)
        for idx in enc_inputs:
            embedded[idx] += 1
        embedded = embedded.unsqueeze(0)  # (1, 10)
        
        output = self.encoder(embedded)  # (1, latent_dim * 2)
        mu = output[:, :self.latent_dim]  # (1, latent_dim)
        logvar = output[:, self.latent_dim:]  # (1, latent_dim)
        return mu, logvar
    
    def sample_z(self, mu, logvar):
        # Sample from N(mu, exp(logvar))
        std = torch.exp(0.5 * logvar)
        eps = torch.randn_like(std)
        z = mu + eps * std
        return z

# Create instances
model = VAEModel(latent_dim=4)
vocab = Vocab()
sequence = "hello"
sample_q = 'max'

ARGS = {
    'model': model,
    'vocab': vocab,
    'sequence': sequence,
    'sample_q': sample_q
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (1, 4) — shape (1, 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
evaluate_nll Ran
ran on a synthesized input
IBM/amp-design/density_modeling.py
code served (permissive licence) · get_code("aee5983583e02518")
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, 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 torch
import math
import torch.nn as nn

torch.manual_seed(0)

# Create a simple variational distribution q(z) with a logpdf method
class SimpleVariationalDistribution(nn.Module):
    def __init__(self, dim):
        super().__init__()
        self.dim = dim
        self.mu = nn.Parameter(torch.randn(dim))
        self.log_var = nn.Parameter(torch.randn(dim))
    
    def logpdf(self, z):
        """Compute log probability density of z under this Gaussian distribution."""
        if isinstance(z, float):
            z = torch.tensor(z, dtype=torch.float32)
        else:
            z = torch.tensor(z, dtype=torch.float32)
        
        # Standard Gaussian log pdf: -0.5 * (z - mu)^2 / var - 0.5 * log(2*pi*var)
        var = (self.log_var).exp()
        diff = z - self.mu
        logpdf_val = -0.5 * ((diff ** 2) / var).sum() - 0.5 * self.dim * math.log(2 * math.pi) - 0.5 * self.log_var.sum()
        return logpdf_val.item()

# Instantiate the variational distribution
q = SimpleVariationalDistribution(dim=4)

# Create points: (mu, log_variance) tensors
# mu has shape (N, D) where N=2 is batch size, D=4 is dimension
mu = torch.randn(2, 4)
lv = torch.randn(2, 4)  # log variance
points = (mu, lv)

ARGS = {
    "q": q,
    "points": points
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple 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
kl_gaussian_sharedmu Ran
ran on a synthesized input
IBM/amp-design/losses.py
code served (permissive licence) · get_code("c5140cfc444d438f")
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 Tensor, 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 torch

torch.manual_seed(0)

# Create small tensors for the function parameters
# mu: mean tensor, shape (batch_size, latent_dim)
# logvar: log variance tensor, shape (batch_size, latent_dim)
batch_size = 2
latent_dim = 4

mu = torch.randn(batch_size, latent_dim)
logvar = torch.randn(batch_size, latent_dim)

ARGS = {
    "mu": mu,
    "logvar": logvar,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, 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
kl_gaussianprior Ran
ran on a synthesized input
IBM/amp-design/losses.py
code served (permissive licence) · get_code("1160cb2411dbed20")
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 Tensor, 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 torch

torch.manual_seed(0)

# The function expects:
# - mu: a tensor that supports ** 2 operation and torch.sum
# - logvar: a tensor that supports .exp(), ** 2 operation, and torch.sum
# Both should be 2D tensors where dim 1 is summed over

batch_size = 2
latent_dim = 4

mu = torch.randn(batch_size, latent_dim)
logvar = torch.randn(batch_size, latent_dim)

ARGS = {
    "mu": mu,
    "logvar": logvar,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, 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
onehot_embed Ran
ran on a synthesized input
IBM/amp-design/models/mutils.py
code served (permissive licence) · get_code("0c9ed4a5c07ec053")
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 Tensor, ndim 2, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# Create a 1D tensor of indices (minibatch of size 2)
# Each index should be valid (in range [0, vocabSize))
hardIx = torch.tensor([0, 3], dtype=torch.long)

# Define vocabulary size
vocabSize = 4

ARGS = {
    "hardIx": hardIx,
    "vocabSize": vocabSize
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 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
prior_logpdf Ran
ran on a synthesized input
IBM/amp-design/density_modeling.py
code served (permissive licence) · get_code("c8a4fc9dea5a533b")
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 float, 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 torch
import math

torch.manual_seed(0)

# Create a small tensor for z
# The function expects z to have shape (D,) where D is the dimensionality
# It accesses z.shape[0] to get D, and computes (z ** 2).sum()
# It also calls .item() on the energy result, so energy must be a scalar tensor
z = torch.randn(4)

ARGS = {
    "z": z
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
sample_from_model Ran
ran on a synthesized input
IBM/amp-design/api.py
code served (permissive licence) · get_code("110055a8042d12f7")
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 dict. 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 torch.nn as nn

torch.manual_seed(0)

# Stand-in Vocab class
class Vocab:
    def __init__(self):
        self.word_map = {0: "the", 1: "cat", 2: "sat", 3: "<pad>"}
    
    def to_word(self, idx, print_special_tokens=True):
        if isinstance(idx, torch.Tensor):
            idx = idx.item()
        word = self.word_map.get(idx, "<unk>")
        if not print_special_tokens and word.startswith("<"):
            return ""
        return word

# Stand-in Model class
class SimpleVAE(nn.Module):
    def __init__(self, hid_size=4, vocab_size=4):
        super().__init__()
        self.hid_size = hid_size
        self.vocab_size = vocab_size
        self.linear = nn.Linear(hid_size, vocab_size)
    
    def generate_sentences(self, n_samples, z=None, c=None, sample_mode='greedy', **kwargs):
        if z is None:
            z = torch.randn(self.hid_size, n_samples)
        if c is None:
            c = torch.randint(0, 2, (1, n_samples))
        
        # Generate sample indices
        if sample_mode == 'beam':
            # Return list of lists of tensors for beam search
            samples = [[torch.tensor([i]) for i in range(n_samples)] for _ in range(n_samples)]
        else:
            # Return list of tensors for greedy
            samples = [torch.tensor(i % self.vocab_size) for i in range(n_samples)]
        
        return samples, z, c

# Create instances
model = SimpleVAE(hid_size=4, vocab_size=4)
vocab = Vocab()
z = torch.randn(4, 2)
c = torch.randint(0, 2, (1, 2))

ARGS = {
    'model': model,
    'vocab': vocab,
    'z': z,
    'c': c,
    'n_samples': 2,
    'print_special_tokens': True,
    'sample_mode': 'greedy'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
soft_embed Ran
ran on a synthesized input
IBM/amp-design/models/mutils.py
code served (permissive licence) · get_code("320fb8d2a66984fa")
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 Tensor, ndim 2, dtype family float, shape (2, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create a small nn.Embedding instance
# vocab_size=5, embedding_dim=4
embed = nn.Embedding(num_embeddings=5, embedding_dim=4)

# Create softIx as a tensor that can be matrix-multiplied with embed.weight
# embed.weight has shape (5, 4)
# softIx should have shape (2, 5) so that softIx @ embed.weight gives (2, 4)
softIx = torch.randn(2, 5)

ARGS = {
    'embed': embed,
    'softIx': softIx
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 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
build_flow Not yet run
unverified (harvested, no recorded run)
IBM/amp-design/models/flow.py
code served (permissive licence) · get_code("f6fcd3d87fd9b7f2")
How it ran

no invocation recorded.

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 2026-11-21.

Repositories linked to this paper

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

Abstract

De novo therapeutic design is challenged by a vast chemical repertoire and multiple constraints, e.g., high broad-spectrum potency and low toxicity. We propose CLaSS (Controlled Latent attribute Space Sampling) - an efficient computational method for attribute-controlled generation of molecules, which leverages guidance from classifiers trained on an informative latent space of molecules modeled using a deep generative autoencoder. We screen the generated molecules for additional key attributes by using deep learning classifiers in conjunction with novel features derived from atomistic simulations. The proposed approach is demonstrated for designing non-toxic antimicrobial peptides (AMPs) with strong broad-spectrum potency, which are emerging drug candidates for tackling antibiotic resistance. Synthesis and testing of only twenty designed sequences identified two novel and minimalist AMPs with high potency against diverse Gram-positive and Gram-negative pathogens, including one multidrug-resistant and one antibiotic-resistant K. pneumoniae, via membrane pore formation. Both antimicrobials exhibit low in vitro and in vivo toxicity and mitigate the onset of drug resistance. The proposed approach thus presents a viable path for faster and efficient discovery of potent and selective broad-spectrum antimicrobials.

For agents

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

get_harvested_code_for_paper("2005.11248")
get_code_for_paper("2005.11248")
have("2005.11248")

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

curl https://syntology.ai/api/ran/2005.11248.json

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

[![Syntology run record](https://syntology.ai/api/ran/2005.11248.svg)](https://syntology.ai/paper/2005.11248)

Connect an agent — have() is free.