TSMDA Help Page

How to use TSMDA


TSMDA is a machine learning-based model that leverages target and symptom information and two robust negative sample selection approaches to accurately predict potential miRNA-disease associations. This model was created based on available known associations in HMDD v2.0.

(A) represents the main page of TSMDA:

  • Users are directed to the submission page when clicking on “Prediction” at the top menu (1).
  • Data used to train and validate the model can be found on the “Data” tab (2) .
  • In case you experience any trouble using TSMDA or if you have any suggestions or comments, please do not hesitate to contact us either via Biosig’s webpage (Subject: TSMDA) by clicking on the “Contact” page (3) or by email.
    • If you are contacting us regarding a job submission, please include its details such as input information and the job identifier.

Submission Page

(B) depicts the submission page:

  • Firstly, users need to provide a set of miRNAs as a file (1A) OR as a single string (1B). Only miRNA IDs in miRBase are acceptable.
  • Similarly, users must submit a set of diseases using the Medical Subject Heading (MeSH) format. This set can be defined using a file (2A) OR a single string (2B). The MeSH IDs are available at MeSH Browser (nih.gov) .
  • Users can fill the email addresses (3) for sending a notice with the result link when TSMDA prediction is done.
  • By clicking on “Predict miRNA-disease associations(4), the provided list of miRNA and disease will be submitted to TSMDA webserver for prediction.

Waiting Page

(C) represents the waiting step:

  • In this step, the disease-association query is being processed by TSMDA. The maximum waiting time is shown for each submission. Users can bookmark this page and come back later to check the results. If a valid email address was filled in on the submission page, a notice with the result link will be received through it.

Results Page

(D) presents the result page for TSMDA.

  • The prediction results are shown as a table (1). The information included miRNA ID, MeSH ID, Disease, Predicted Association, Association Confidence, and Evidence in MNDR and dbDEMC databases in terms of PubMed IDs (PMIDs).
  • PMIDs are provided only if miRNA-disease pairs have been confirmed to be associated in MNDR and/or dbDEMC databases.
  • Users have the options to run another prediction (2), download the result in a comma-separated values (CSV) format (3), and/or download the error logs (4). Frequently errors are found in the format of miRNAs and diseases (i.e., in the respective miRNA and/or MeSH IDs).

Supplementary Information: MeSH ID Retrieval Tutorial

Download help page as a PDF