ASCEND

BY NTHRYS

NTHRYS › Internships › Ai Drug Repurposing

SMILES-Based Drug Representation Learning

Ai Drug Repurposing

SMILES-Based Drug Representation Learning

Focused area

Variant

Pay · Join

Step 3 of 5Choose focused area

Field

▾

Category

▾

Focused area

▾

You might prefer

AI Knowledge Graph Drug Repurposing Research Machine Learning Omics-Driven Repurposing Studies AI Drug Repurposing for Rare Disease Research Deep Learning for Drug Side Effect Repurposing AI Literature Mining for Repurposing Signals AI Polypharmacology-Based Repurposing Research AI Patient Data-Driven Drug Repurposing Research AI Pandemic Drug Repurposing Response Research AI Network Medicine Drug Repurposing Research AI Combination Therapy Repurposing Studies AI Molecular Docking Drug Repurposing Graph Neural Networks Drug Target Prediction Transformer Models Biomedical Text Mining AI Cancer Drug Repurposing Pipeline Federated Learning Drug Discovery Networks AI Antimicrobial Resistance Drug Repurposing Clinical Trial Data Mining Repurposing AI Neurodegenerative Disease Repurposing Reinforcement Learning Drug Optimization Protein Structure AI Drug Repurposing AI Cardiovascular Drug Repurposing Research Chemical Similarity Network Drug Discovery Multi-Task Learning Disease Prediction AI Inflammatory Disease Drug Repurposing Sequence-Based Protein Target Discovery Metabolomics Data Integration Drug Repurposing AI Infectious Disease Drug Repurposing Transfer Learning Drug Response Prediction AI Autoimmune Disease Drug Repurposing Gene Expression Drug Response Modeling AI Metabolic Disease Drug Repurposing Attention Mechanisms Drug Target Ranking AI Pulmonary Disease Drug Repurposing Pharmacophore Matching AI Drug Discovery AI Renal Disease Drug Repurposing Synthetic Data Generation Drug Training AI Gastrointestinal Disease Drug Repurposing Causal Inference Drug Effect Attribution AI Ophthalmologic Drug Repurposing Temporal Disease Progression Drug Modeling AI Dermatological Drug Repurposing Research Cross-Species Drug Efficacy Translation AI Orthopedic Disease Drug Repurposing Explainable AI Drug Repurposing Models AI Endocrine Disease Drug Repurposing Zero-Shot Learning Drug Repurposing AI Psychiatric Drug Repurposing Studies Variational Autoencoder Drug Discovery AI Biomarker Discovery Drug Repurposing

SMILES-Based Drug Representation Learning

Intern will create deep learning models that learn latent representations of drug structures from SMILES notation for improved repurposing predictions.

The focused areas below are internship topics in varied working formats. Pick one, then choose your internship type, mode… Read more

🎓 TYPE

🌐 MODE

⏱

📚 Academic: Thesis & PPT assistance included🧪 Tech: Master the protocols hands-on📝 Research > 3 months: Publication co-authorship in a Scopus-indexed journal

🔍

Showing 1–10 of 10

Molecular Graph Neural Networks for SMILES Encoding Optimization

This research investigates how graph neural networks can learn superior representations of chemical structures from SMILES strings by capturing molecular topology and atom-bond relationships. The work produces novel encoding architectures that improve predictive accuracy for drug-target binding and repurposing candidate identification.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Transfer Learning Paradigms Across SMILES Chemical Space Domains

This study explores how pre-trained models on large SMILES datasets can be adapted to predict pharmacological properties in understudied disease areas. Results enable faster identification of repurposing candidates by leveraging knowledge from well-characterized chemical spaces.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Contrastive Learning for Self-Supervised SMILES Representation Discovery

This research develops contrastive frameworks that learn meaningful SMILES representations without labeled data by comparing molecular augmentations and structural variations. The approach generates robust molecular embeddings that enhance drug repurposing prediction accuracy without extensive annotation.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Interpretable Feature Extraction from SMILES for Drug Property Prediction

This investigation identifies which chemical substructures and SMILES tokens most strongly influence predictions of therapeutic efficacy and off-target effects. The interpretable features reveal molecular design principles that guide rational drug repurposing strategies.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Attention-Based SMILES Tokenization for Multi-Task Pharmacological Modeling

This research develops attention mechanisms that dynamically weight different molecular fragments in SMILES representations for simultaneous prediction of multiple pharmacological endpoints. The method produces unified molecular embeddings that improve repurposing accuracy by capturing polypharmacology patterns.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Generative SMILES Models for Novel Drug Candidate Space Expansion

This study creates generative models that synthesize novel SMILES strings with desired pharmacological properties while maintaining synthetic accessibility and safety constraints. The approach expands the chemical space explored during repurposing to discover previously unconsidered therapeutic candidates.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

SMILES Augmentation Strategies for Robust Deep Learning Model Training

This research develops systematic approaches for augmenting SMILES representations through canonical transformations and structural variations to improve model generalization. The augmentation techniques produce more reliable predictions that reduce false positives in repurposing candidate screening.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Molecular Similarity Metrics from SMILES Embeddings for Repurposing Prediction

This investigation designs novel distance metrics in SMILES embedding spaces that better capture functional and structural drug similarity relevant to disease targets. The metrics enable more accurate neighbor-based repurposing predictions by reflecting true pharmacological relatedness.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Heterogeneous Graph Networks Integrating SMILES with Biomedical Knowledge

This research constructs heterogeneous graphs combining SMILES-derived molecular representations with protein interactions, disease associations, and clinical outcomes. The integrated approach produces holistic drug representations that identify repurposing opportunities by connecting chemistry to clinical biology.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →

Uncertainty Quantification in SMILES-Based Drug Efficacy Predictions

This study develops Bayesian and ensemble methods for quantifying prediction uncertainty in SMILES-based drug repurposing models to identify high-confidence candidates. The uncertainty estimates enable prioritization of repurposing leads for experimental validation with reduced false discovery rates.

Academic (A)Tech (T)Research (R)

14 more durationsView Titles →