# Database for comparing two samples

**URL:** <https://scientistsolutions.discourse.group/t/database-for-comparing-two-samples/738>\
**Category:** Sequence Analysis\
**Created:** [January 16, 2025, 6:55am UTC](https://scientistsolutions.discourse.group/t/database-for-comparing-two-samples/738 "2025-01-16T06:55:30Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![HarmanAulakh](https://avatars.discourse-cdn.com/v4/letter/h/a87d85/32.png) [@HarmanAulakh](https://scientistsolutions.discourse.group/u/HarmanAulakh)\
**Post date:** [January 16, 2025, 6:55am UTC](https://scientistsolutions.discourse.group/t/database-for-comparing-two-samples/738/1 "2025-01-16T06:55:31Z")

</div>

Hi All,

I am working on HT-MS. I want to compare MS data(proteomic) from two samples/experiments.

Any suggestions are very welcome

---

<div class="post-metadata">

**Author:** ![ARGERINE](https://avatars.discourse-cdn.com/v4/letter/a/8491ac/32.png) [@ARGERINE](https://scientistsolutions.discourse.group/u/ARGERINE)\
**Post date:** [June 16, 2025, 4:24pm UTC](https://scientistsolutions.discourse.group/t/database-for-comparing-two-samples/738/2 "2025-06-16T16:24:57Z")

</div>

Harman In order to compare High Throughput MS data from different samples, begin by organizing protein IDs and intensity values (from tools like MaxQuant or Proteome Discoverer) into a table. Normalize the data using log2 transformation and total or median normalization. Perform statistical analysis (t-test or non-parametric tests) and adjust for multiple testing using FDR. Identify differentially expressed proteins based on fold-change (e.g., ±1.5) and significance (e.g., p ≤ 0.05). Visualize results using volcano plots and perform PCA or clustering to assess sample separation. For functional insights, analyze significant proteins using DAVID, g:Profiler, or Metascape to extract GO terms and pathway associations. Use STRING-db or Cytoscape to explore interaction networks. Optionally, apply machine learning (e.g., random forest) to identify key discriminative proteins. This approach helps you uncover biological meaning from proteomic differences and highlight potential pathogenic or biomarker proteins between conditions. Tools like Perseus, R (limma, MSstats), or Python can support the full analysis pipeline.

---

<div class="post-metadata">

**Author:** ![HarmanAulakh](https://avatars.discourse-cdn.com/v4/letter/h/a87d85/32.png) [@HarmanAulakh](https://scientistsolutions.discourse.group/u/HarmanAulakh)\
**Post date:** [June 20, 2025, 2:13pm UTC](https://scientistsolutions.discourse.group/t/database-for-comparing-two-samples/738/3 "2025-06-20T14:13:47Z")

</div>

Thank you for helping and for guiding.