Long non-coding RNAs (lncRNAs) have been proven to be implicated in the pathogenesis of various diseases. Multiple studies have demonstrated that small molecule drugs can modify lncRNA expression, which suggests a promising therapeutics for human diseases. Here, we constructed a comprehensive query and analysis platform named D-lnc for dissecting the influence of drugs on lncRNA expression. We retrieved the experimentally validated effects (up-regulation or down-regulation) of drugs on lncRNA expression from published papers in PubMed. In addition, we comprehensively screened the Connectivity Map Database (cMap) and Gene Expression Omnibus Database (GEO) to obtain the drug-perturbed gene expression profiles. Through probe re-annotation of microarray data, we identified the putative drug affected lncRNAs.

There are three datasets that contain the associations of lncRNAs and drugs:
1. Validated dataset: manually curated the effects (up-regulation or down-regulation) of drugs on lncRNA expression from literatures.
2. cMap dataset: re-annotated the probes of microarray data in cMap to obtain the drug-perturbed lncRNA expression profiles. The significantly differentially expressed lncRNAs between drug-treated samples and control samples were considered as drug-affected lncRNAs.
3. GEO dataset: re-annotated the probes of microarray data in GEO to obtain the drug-perturbed lncRNA expression profiles. We comprehensively searched the GEO to find the microarray data with drug treatment. The significantly differentially expressed lncRNAs between drug-treated samples and control samples were considered as drug-affected lncRNAs.

Finally, we proved that lncRNAs with similar sequences were more likely to be influenced by the same small molecule, and small molecules with similar structures were more likely to regulate the same lncRNA. Based on this hypothesis, we developed an on-line analysis module to predict the potential related drugs or lncRNAs according to users' inputted lncRNA sequences or drug structures.
1. Analysis By lncRNA Sequence: the blast-2.7.1+ is used to calculate the sequence similarity of two lncRNAs. The lncRNAs with BLAST E-value < 10 are considered as lncRNAs with similar sequences.
2. Analysis By Small Molecule Structure: the maximum common substructure (MCS) algorithm is used to calculate the structure similarity of two drugs. The drugs with Tanimoto Identity (Sim) > 0.6 are considered as drugs with similar structures.