DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction

DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction | Signal Canvas | ScienceToStartup

Page Freshness

Signal Canvas proof surface

Canonical route: /signal-canvas/dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction

stale

Proof freshness: stale
Proof status: unverified
Display score: 8/10
Last proof check: 2026-03-30
Score updated: 2026-04-02
Score fresh until: 2026-05-02
References: 68
Source count: 3
Coverage: 50%

This page is showing the last landed evidence receipt and score bundle because the latest proof data is outside the freshness window.

Agent Handoff

Canonical ID dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction | Route /signal-canvas/dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction

REST example

curl https://sciencetostartup.com/api/v1/agent-handoff/signal-canvas/dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction

MCP example

{
  "tool": "search_signal_canvas",
  "arguments": {
    "mode": "paper",
    "paper_ref": "dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction",
    "query_text": "Summarize DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction"
  }
}

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction",
  "normalized_query": "2603.26114",
  "route": "/signal-canvas/dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction",
  "paper_ref": "dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}

Evidence Receipt

Route status: building

Claims: 12

References: 68

Proof: Verification pending

Freshness state: computing

Source paper: DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction

PDF: https://arxiv.org/pdf/2603.26114v1

Source count: 3

Coverage: 50%

Last proof check: 2026-03-30T21:51:22.698Z

Signal Canvas receipt window

Watch and verify: DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction

/buildability/dpd-cancer-explainable-graph-based-deep-learning-for-small-molecule-anti-cancer-activity-prediction

Watchwatch

Subject: DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction

Verdict

Watch

Verdict is Watch because viability or proof quality is intermediate and should be re-evaluated before execution.

Preparing verified analysis

GitHub Code Pulse

No public code linked for this paper yet.

Claim map

Strong 12Mixed 0Weak 0

Evidencepartial
DPD-Cancer demonstrated superior performance, achieving an Area Under ROC Curve (AUC) of up to 0.87 on strictly partitioned NCI60 data
Implicationpartial
Directly stated in the abstract and supported by performance tables.
Verificationpartial
partial
Evidencepartial
and up to 0.98 on ACLPred/MLASM datasets.
Implicationpartial
Directly stated in the abstract and supported by performance tables.
Verificationpartial
partial
Evidencepartial
For pGI50 prediction across 10 cancer types and 73 cell lines, the model achieved Pearson's correlation coefficients of up to 0.72 on independent test sets.
Implicationpartial
Directly stated in the abstract and supported by performance tables.
Verificationpartial
partial
Evidencepartial
we introduce DPD-Cancer, a deep learning method based on a Graph Attention Transformer (GAT) framework. It is designed for small molecule anti-cancer activity classification and the quantitative prediction of cell-line specific responses
Implicationpartial
The abstract explicitly states the method used and its purpose.
Verificationpartial
partial
Evidencepartial
Furthermore, DPD-Cancer provides explainability by leveraging the attention mechanism to identify and visualise specific molecular substructures
Implicationpartial
The abstract and analysis explicitly mention the explainability feature and how it's achieved.
Verificationpartial
partial
Evidencepartial
Reliance on the NCI60 dataset may limit generalizability across other molecular sets.
Implicationpartial
This is stated as a caveat in the analysis section, implying a limitation.
Verificationpartial
partial
Evidencepartial
It uses chemistry-aware data partitioning to ensure robust evaluation on novel chemotypes
Implicationpartial
Mentioned in the analysis section as a methodology for robust evaluation.
Verificationpartial
partial
Evidencepartial
DPD-Cancer demonstrated superior performance, achieving an Area Under ROC Curve (AUC) of up to 0.87 on strictly partitioned NCI60 data
Implicationpartial
The abstract explicitly states this performance metric and dataset.
Verificationpartial
partial
Evidencepartial
and up to 0.98 on ACLPred/MLASM datasets.
Implicationpartial
The abstract explicitly states this performance metric and dataset.
Verificationpartial
partial
Evidencepartial
For pGI50 prediction across 10 cancer types and 73 cell lines, the model achieved Pearson's correlation coefficients of up to 0.72 on independent test sets.
Implicationpartial
The abstract explicitly states this performance metric and context.
Verificationpartial
partial
Evidencepartial
we introduce DPD-Cancer, a deep learning method based on a Graph Attention Transformer (GAT) framework.
Implicationpartial
The abstract clearly states the underlying deep learning method.
Verificationpartial
partial
Evidencepartial
Furthermore, DPD-Cancer provides explainability by leveraging the attention mechanism to identify and visualise specific molecular substructures, offering actionable insights for lead optimisation.
Implicationpartial
The abstract explicitly mentions the explainability feature and how it's achieved.
Verificationpartial
partial

Author intelligence and commercialization panels stay hidden until the proof receipt is verified, cites at least 3 references, includes at least 2 sources, and clears 50% coverage. The paper narrative and citation surfaces remain public while verification is pending.

DPD-Cancer: Explainable Graph-based Deep Learning for Small Molecule Anti-Cancer Activity Prediction

Use Signal Canvas as the narrative proof surface

Use this Signal Canvas via API or MCP

Signal Canvas proof surface