Problems of downloaded data "LINCS Small Molecules Metadata"

[liuxiawei]

Hello, I downloaded the LINCS Small Molecules Metadata data from SigCom LICNS (Download page, LINCS Small Molecules Metadata, https://s3.amazonaws.com/lincs-dcic/sigcom-lincs-metadata/LINCS_small_molecules.tsv). However, I'm having trouble understanding some of the information, such as the "sig_count" field. I couldn't find any explanation for it on Google, SigCom LINCS, or the CLUE. Could you please help me understand what this means or guide me to a resource where I can find an explanation? Thank you for your assistance!

[AviMaayan]

pert_name: BRD ID is the unique Broad Institute ID followed by a common name
target: Entrez Gene Symbols of the known targets for the drug
moa: Known Mechanisms of Action for the drug
canonical_smiles: Chemical representation of the compound using SMILES annotations
inchi_key: Chemical representation of the compound using INCHI keys compound_aliases: Other names for the compound
sig_count: Number of signatures for the compound in SigCom LINCS. Signatures are defined as drug treatments of a specific cell line, at a specific concentration, and the time point when gene expression was measured after drug treatment.

[liuxiawei]

Thanks, AviMaayan! It helps a lot ! But I found sig_count is not consistent with the web. Like compound losartan, It was 395 in downloaded file (Download page, LINCS Small Molecules Metadata, https://s3.amazonaws.com/lincs-dcic/sigcom-lincs-metadata/LINCS_small_molecules.tsv). But in web meta search , only 328 Signature found by searching losartan. Could it be that my understanding is incorrect? Or is the inconsistency in the data due to version or other issues?

[jeevangelista]

Hi @liuxiawei, Sorry for the confusion, I've updated the small molecules files to reflect the proper counts

[liuxiawei]

Thanks @jeevangelista ! It changes and looks like more appropriate. I still find the compound losartan is not right. There is still a difference of 1 between their results (328 in Web, and 327 in new tsv file in total). I suspect there may still be some issues, but it does not have a significant impact for me . I am glad that I was able to confirm that my understanding is correct. Thank you for all your work.

[jeevangelista]

Hi @liuxiawei, the extra signature is from the consensus signatures that we built, which is a completely different dataset of consensus chemical perturbation signatures

[liuxiawei]

Thank you for your response. I now understand, and I believe that having a wiki for this database would definitely attract more users. Once again, thank you!