# Prodigy cannot predict binding affinities for protein-protein by GRAMM-X docking model.

**URL:** <https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154>\
**Category:** PRODIGY\
**Created:** [March 6, 2023, 9:15am UTC](https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154 "2023-03-06T09:15:33Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Wutthipat1](https://avatars.discourse-cdn.com/v4/letter/w/e495f1/32.png) [@Wutthipat1](https://ask.bioexcel.eu/u/Wutthipat1)\
**Post date:** [March 6, 2023, 9:15am UTC](https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154/1 "2023-03-06T09:15:33Z")

</div>

I submitted my PDB file and got the following error.

# ===========================================================================

command: /usr/local/bin/prodigy model.1\_1.pdb --selection A C --temperature 25.0 --contact\_list --pymol\_selection  
exit status: 1  
log:  
[+] Reading structure file: /data/runs/3QaRoADl/model.1\_1.pdb  
[+] Parsed structure file model.1\_1 (1 chains, 68 residues)  
Traceback (most recent call last):  
File “/usr/local/bin/prodigy”, line 8, in   
sys.exit(main())  
File “/usr/local/lib/python3.10/dist-packages/prodigy/predict\_IC.py”, line 389, in main  
prodigy.predict(  
File “/usr/local/lib/python3.10/dist-packages/prodigy/predict\_IC.py”, line 152, in predict  
self.ic\_network = calculate\_ic(  
File “/usr/local/lib/python3.10/dist-packages/prodigy/predict\_IC.py”, line 65, in calculate\_ic  
raise ValueError(“No contacts found for selection”)  
ValueError: No contacts found for selection

I have checked the PDB files with PDB tools and the result is below.

 ![image](https://europe1.discourse-cdn.com/flex013/uploads/bioexcel/original/1X/cca12f691fb20145b1d5f88549361e1ac2b5fe40.png)

---

<div class="post-metadata">

**Author:** ![honoratorv](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/honoratorv/32/672_2.png) [@honoratorv](https://ask.bioexcel.eu/u/honoratorv)\
**Post date:** [March 6, 2023, 10:15am UTC](https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154/2 "2023-03-06T10:15:00Z")

</div>

Could you visually inspect the complex with (chimera or pymol or vmd) and confirm that chain A is in contact with chain B?

---

<div class="post-metadata">

**Author:** ![Wutthipat1](https://avatars.discourse-cdn.com/v4/letter/w/e495f1/32.png) [@Wutthipat1](https://ask.bioexcel.eu/u/Wutthipat1)\
**Post date:** [March 6, 2023, 8:41pm UTC](https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154/3 "2023-03-06T20:41:16Z")

</div>

I did initial confirmation with Pymol and found that the two proteins actually contracted.

 ![image](https://europe1.discourse-cdn.com/flex013/uploads/bioexcel/original/1X/02e9a2ac6cffff66ba95a5ddde1b82b4b98d27ac.png)

---

<div class="post-metadata">

**Author:** ![amjjbonvin](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/amjjbonvin/32/23_2.png) [@amjjbonvin](https://ask.bioexcel.eu/u/amjjbonvin)\
**Post date:** [March 6, 2023, 10:03pm UTC](https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154/4 "2023-03-06T22:03:07Z")

</div>

In your prodigy submission you defined chains A and C… are these correct? From the pymol color coding I would think you are dealing with chains A and B

---

<div class="post-metadata">

**Author:** ![honoratorv](https://dub1.discourse-cdn.com/flex013/user_avatar/ask.bioexcel.eu/honoratorv/32/672_2.png) [@honoratorv](https://ask.bioexcel.eu/u/honoratorv)\
**Post date:** [March 7, 2023, 11:11am UTC](https://ask.bioexcel.eu/t/prodigy-cannot-predict-binding-affinities-for-protein-protein-by-gramm-x-docking-model/4154/5 "2023-03-07T11:11:34Z")

</div>

The chains are colored by order (first green, second cyan, etc) not by it’s id.

@Wutthipat1 thanks for checking, if possible could you please send me the input structure for me to run some tests? \> [rvhonorato@protonmail.com](mailto:rvhonorato@protonmail.com)