pfindstudio / pdeep Goto Github PK
View Code? Open in Web Editor NEWpDeep: Predicting MS/MS Spectra of Peptides with Deep Learning
License: Other
pDeep: Predicting MS/MS Spectra of Peptides with Deep Learning
License: Other
Hello,
First of all, I want to thank pFinders for providing so many convenient and practical software. You are really a great team.
Recently, I want to use the pDeep2 software to perform some verification on some suspected PSMs in our previous identification results. I encountered some problems in the process of doing it.
As a measure of similarity between predicted spec and real spec, does PCC have a threshold to distort true positives and false positives? For example, we will use the Andromeda score >100 as the threshold for MaxQuant identification in some cases.
If there is no such threshold yet, I may need to do some statistical testing to find a way out.
I would like to know whether the ProteomeTools data was used in the model training process corresponding to the-i Lumos, -e 0.30
parameter in the current version (model-180921). This determines whether I can use this type of data from ProteomeTools as my data set for statistical testing.
Thank you.
Zhang Le
I saw in plabel files where neutral loss ions are included but I didn't see the prediction result for those ions. Are these ions considered in this version?
Dear pDeep Team,
I have two questions concerning pDeep2:
A) How can we add other modifications (like e.g. Propionyl)?
B) Is there a possibility to directly compare/calculate the Pearson Correlation between measured spectra (..mgf file) and the predictions?
Regards,
Bart
Is there a pre-trained model for EThcD? When I set mod_config = fconfig.EThcD_Config() the predictions were the same as with HCD.
In pDeep2 paper, we found that the PCC90 of the test data set, HPM: fetal gut
, was quite high (87.5% in pDeep2 paper). Then we fixed a bug in https://github.com/pFindStudio/pDeep/blob/master/pDeep2/model/featurize.py, at line 252
252 if instrument not in self.instrument_feature:
253 inst_feature = self.instrument_feature[instrument.lower()]
254 else:
255 inst_feature = self.instrument_feature['unknown']
The correct code:
252 if instrument in self.instrument_feature:
After this bug was fixed, PCC90 of HPM: fetal gut
decreased quite a lot, but PCC90 of PT25
became much better. To reproduce the results in pDeep2 paper, we could just modify the code at line 252 back.
There is no such a bug in pDeep3 repo, because it is already fixed in the initial commit.
Hi, I am not able to use pDeep for ETD data. pDeep only generating b and y ion. Can you please guide me on how I can generate c and z ion for ETD data.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.