# Data issue with PRISM drug repurposing dataset

**URL:** <https://forum.depmap.org/t/data-issue-with-prism-drug-repurposing-dataset/3124>\
**Category:** Report an Issue\
**Tags:** compound-screens, data\
**Created:** [March 8, 2024, 2:08pm UTC](https://forum.depmap.org/t/data-issue-with-prism-drug-repurposing-dataset/3124 "2024-03-08T14:08:18Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Karen\_Wong](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.depmap.org/karen_wong/32/1062_2.png) [@Karen\_Wong](https://forum.depmap.org/u/Karen_Wong)\
**Post date:** [March 8, 2024, 2:08pm UTC](https://forum.depmap.org/t/data-issue-with-prism-drug-repurposing-dataset/3124/1 "2024-03-08T14:08:18Z")

</div>

Hi,

I’ve been analyzing compound screening data from the PRISM Repurposing Secondary Screen and have encountered a potential data or technical artifact. For example, with the compound SN38, I noticed that nearly all cell lines exhibit unexpectedly high viability values at the dose of 0.0024µM. This pattern appears consistent across cell lines derived from various tissues, suggesting a systemic anomaly. Could it be that the viability values for the two lowest doses have been inadvertently swapped? Alternatively, there might have been a malfunction with the automatic dispenser, possibly leading to incorrect dispensation of the 0.0024µM dose. Thanks!

---

<div class="post-metadata">

**Author:** ![dllahr](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.depmap.org/dllahr/32/23_2.png) [@dllahr](https://forum.depmap.org/u/dllahr)\
**Post date:** [March 8, 2024, 5:16pm UTC](https://forum.depmap.org/t/data-issue-with-prism-drug-repurposing-dataset/3124/2 "2024-03-08T17:16:42Z")

</div>

This is interesting / important to look into. When you say the viability values are unexpectedly high, is that high with respect to the DMSO? I.e. how do you define high? Or another way of asking - how much higher are these values than the DMSO values?

---

<div class="post-metadata">

**Author:** ![Karen\_Wong](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.depmap.org/karen_wong/32/1062_2.png) [@Karen\_Wong](https://forum.depmap.org/u/Karen_Wong)\
**Post date:** [March 8, 2024, 5:25pm UTC](https://forum.depmap.org/t/data-issue-with-prism-drug-repurposing-dataset/3124/3 "2024-03-08T17:25:11Z")

</div>

They are unexpectedly higher than the first and third doses in a very consistent manner. Attached are some example plots. I was browsing around [here](https://depmap.org/portal/compound/SN-38?tab=dose-curves) and almost all of them seem to have the described pattern.

 ![image](https://canada1.discourse-cdn.com/flex035/uploads/depmap/original/2X/7/7875d42a8d6f7a1b7dc1d82c19b9a4448e61a588.png)  
 ![image](https://canada1.discourse-cdn.com/flex035/uploads/depmap/original/2X/f/fe1c23b7129e66ee89d957b8280cad4c04a8f6a0.png)  
 ![image](https://canada1.discourse-cdn.com/flex035/uploads/depmap/original/2X/d/d25c9a1b233edf7735a9a81b798f0b0cbc81cf33.png)