SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.21550 · 2026

Extending Sequence Length is Not All You Need: Effective Integration of Multimodal Signals for Gene Expression Prediction

Bing Su, Zhao Yang, Jiwei Zhu, Ying Ba, Chuan Cao, Yi Duan

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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 0 have no contract verdict; 5 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/2602.21550.json · badge: /api/ran/2602.21550.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
yangzhao1230/Prism canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
cage_pred_collate_fn Ran
ran; our contract draft was wrong, not the code
yangzhao1230/Prism/src/dataloaders/genomics.py
code served (permissive licence) · get_code("52c9c163b1a53a23")
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 torch
from torch_geometric.data import Data, Batch

torch.manual_seed(0)

# Create a minimal batch for cage_pred_collate_fn
# The function expects: batch = list of tuples (graph_data, signals, token_dna)
# where:
#   - graph_data is a torch_geometric.data.Data object
#   - signals is a 1D tensor
#   - token_dna is a 1D tensor

# Create sample graph data objects
graph1 = Data(
    x=torch.randn(5, 4),  # 5 nodes, 4 features
    edge_index=torch.tensor([[0, 1, 2], [1, 2, 3]], dtype=torch.long),
    y=torch.tensor([0])
)

graph2 = Data(
    x=torch.randn(6, 4),  # 6 nodes, 4 features
    edge_index=torch.tensor([[0, 1, 2, 3], [1, 2, 3, 4]], dtype=torch.long),
    y=torch.tensor([1])
)

# Create signals tensors (1D tensors of size 4)
signals1 = torch.randn(4)
signals2 = torch.randn(4)

# Create token_dna tensors (1D tensors of size 4)
token_dna1 = torch.randint(0, 5, (4,))
token_dna2 = torch.randint(0, 5, (4,))

# Construct the batch as a list of tuples
batch = [
    (graph1, signals1, token_dna1),
    (graph2, signals2, token_dna2)
]

ARGS = {
    "batch": batch
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torch_geometric — 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
cross_entropy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yangzhao1230/Prism/caduceus/modeling_caduceus.py
pointer only (licence: MIT); the stored copy get_code serves, from divelab/AIRS, is pointer only (licence: GPL-3.0) · get_code("83de732cfd3e479f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from divelab/AIRS, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

fftconv_h3_ref Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yangzhao1230/Prism/src/ops/fftconv.py
code served (permissive licence) · get_code("5d11b94982686ba5")
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 3, 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
from einops import rearrange

torch.manual_seed(0)

# Small tensor dimensions
batch = 2
head_dim = 1
num_heads = 3
d1 = head_dim
d2 = head_dim
seqlen = 4

# k: shape (batch, h*d1, seqlen)
k = torch.randn(batch, num_heads * d1, seqlen)

# ssm_kernel: shape (seqlen,) or similar for FFT
ssm_kernel = torch.randn(seqlen)

# D: shape (num_heads * d2,) for the skip connection
D = torch.randn(num_heads * d2)

# q: shape (batch, h*d1, seqlen)
q = torch.randn(batch, num_heads * d1, seqlen)

# v: shape (batch, h*d2, seqlen)
v = torch.randn(batch, num_heads * d2, seqlen)

# head_dim parameter
head_dim_param = 1

# ssm_kernel_rev: optional, can be None or a tensor
ssm_kernel_rev = None

ARGS = {
    'k': k,
    'ssm_kernel': ssm_kernel,
    'D': D,
    'q': q,
    'v': v,
    'head_dim': head_dim_param,
    'ssm_kernel_rev': ssm_kernel_rev,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float — no output shape is stated: observe_signature.py could not observe it (import:OSError)
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
fftconv_ref Ran
ran; our contract draft was wrong, not the code
yangzhao1230/Prism/src/ops/fftconv.py
pointer only (licence: MIT); the stored copy get_code serves, from divelab/AIRS, is pointer only (licence: GPL-3.0) · get_code("741eb38fcf150c88")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from divelab/AIRS, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

weighted_cross_entropy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yangzhao1230/Prism/caduceus/modeling_caduceus.py
pointer only (licence: MIT); the stored copy get_code serves, from divelab/AIRS, is pointer only (licence: GPL-3.0) · get_code("2ef47d00e83953ae")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from divelab/AIRS, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is 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, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 0 later) and 1 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
cross_entropy 6 other papers
6 dated earlier
2305.04120 (2023-05)
A Latent Diffusion Model for Protein Structure Generation
2403.03234 (2024-03)
Caduceus: Bi-Directional Equivariant Long-Range DNA Sequence Modeling
2412.10193 (2024-12)
Simple Guidance Mechanisms for Discrete Diffusion Models
and 3 more not named here
weighted_cross_entropy 6 other papers
6 dated earlier
2305.04120 (2023-05)
A Latent Diffusion Model for Protein Structure Generation
2403.03234 (2024-03)
Caduceus: Bi-Directional Equivariant Long-Range DNA Sequence Modeling
2412.10193 (2024-12)
Simple Guidance Mechanisms for Discrete Diffusion Models
and 3 more not named here
fftconv_ref 4 other papers
4 dated earlier
2302.10866 (2023-02)
Hyena Hierarchy: Towards Larger Convolutional Language Models
2305.04120 (2023-05)
A Latent Diffusion Model for Protein Structure Generation
2307.08423 (2023-07)
Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
and 1 more not named here
fftconv_h3_ref 1 other paper
1 dated earlier
2403.03234 (2024-03)
Caduceus: Bi-Directional Equivariant Long-Range DNA Sequence Modeling

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 expression prediction, which predicts mRNA expression levels from DNA sequences, presents significant challenges. Previous works often focus on extending input sequence length to locate distal enhancers, which may influence target genes from hundreds of kilobases away. Our work first reveals that for current models, long sequence modeling can decrease performance. Even carefully designed algorithms only mitigate the performance degradation caused by long sequences. Instead, we find that proximal multimodal epigenomic signals near target genes prove more essential. Hence we focus on how to better integrate these signals, which has been overlooked. We find that different signal types serve distinct biological roles, with some directly marking active regulatory elements while others reflect background chromatin patterns that may introduce confounding effects. Simple concatenation may lead models to develop spurious associations with these background patterns. To address this challenge, we propose Prism, a framework that learns multiple combinations of high-dimensional epigenomic features to represent distinct background chromatin states and uses backdoor adjustment to mitigate confounding effects. Our experimental results demonstrate that proper modeling of multimodal epigenomic signals achieves state-of-the-art performance using only short sequences for gene expression prediction.

For agents

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

get_harvested_code_for_paper("2602.21550")
get_code_for_paper("2602.21550")
have("2602.21550")

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

curl https://syntology.ai/api/ran/2602.21550.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.21550.svg)](https://syntology.ai/paper/2602.21550)

Connect an agent — have() is free.