iProphet for Leucine/isoleucine IDs

[Chih-Chiang Tsou]

Dear TPP developers,

I have a question regarding how iProphet handles ambiguous I/L IDs from different search engine. I am using X!Tandem, Comet, and MSGF+ and combine the results using iProphet, and noticed that the three search engines have different rules ranking I/L hits.

So how does iProphet handle such case if different search engines give different I/L variant as top ranking PSM? Does iProphet threat different I/L variant peptide as different IDs and therefore penalizing the final probability score, or iProphet considers them as the same ID?

Thanks,

Chih-Chiang

[David Shteynberg]

Hello Chih-Chiang,

iProphet will convert all Isoleucines reported to Leucines for comparison purposes.  Have you observed an error in your analyses?

-David

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discuss+unsubscribe@googlegroups.com.
To post to this group, send email to spctools-discuss@googlegroups.com.
Visit this group at https://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/d/optout.

[Chih-Chiang Tsou]

Hi David,

Thank you for the quick reply. The three search engines I am using they rank I/L variant peptides inconsistently. Some ranks Leucine-version as rank 1, some ranks Isoleucine-version as rank 1.

Then in iProphet pepXML file I observed that for each spectrum iProphet picks rank 1 hit info from either one of the three search results (the one with largest PeptideProphet probability?) to report in iProphet PepXML. Therefore for some spectrum, Leucine-version is reported, but sometimes Isoleucine-version gets reported.

I was concerned about whether the inconsistent ranking would affect iProphet scoring, but then as you mentioned, iProphet converts I/L variant and treat them as the same sequence therefore it should be fine as far as scoring. Just in the end when we read iProphet result we need to also treat I/L variants properly.  Am I understanding it correctly?

Best,

Chih-Chiang

On Tue, Jun 12, 2018 at 12:26 PM David Shteynberg <David.Sh...@systemsbiology.org> wrote:

Hello Chih-Chiang,

iProphet will convert all Isoleucines reported to Leucines for comparison purposes.  Have you observed an error in your analyses?

-David

On Tue, Jun 12, 2018 at 9:06 AM, Chih-Chiang Tsou <chihchi...@gmail.com> wrote:

Dear TPP developers,

I have a question regarding how iProphet handles ambiguous I/L IDs from different search engine. I am using X!Tandem, Comet, and MSGF+ and combine the results using iProphet, and noticed that the three search engines have different rules ranking I/L hits.

So how does iProphet handle such case if different search engines give different I/L variant as top ranking PSM? Does iProphet threat different I/L variant peptide as different IDs and therefore penalizing the final probability score, or iProphet considers them as the same ID?

Thanks,

Chih-Chiang

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discu...@googlegroups.com.
To post to this group, send email to spctools...@googlegroups.com.

Visit this group at https://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "spctools-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/spctools-discuss/VDEKdnL5iEQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to spctools-discu...@googlegroups.com.
To post to this group, send email to spctools...@googlegroups.com.

[David Shteynberg]

Yes your understanding of the issue is spot on.  iProphet converts everything to L for comparison, then reports the top hit from the highest scoring search engine as reported by the search engine.  Post processing has to take care of how to treat the variants.

Cheers,

-David

On Tue, Jun 12, 2018 at 11:53 AM, Chih-Chiang Tsou <chihchi...@gmail.com> wrote:

Hi David,

Thank you for the quick reply. The three search engines I am using they rank I/L variant peptides inconsistently. Some ranks Leucine-version as rank 1, some ranks Isoleucine-version as rank 1.

Then in iProphet pepXML file I observed that for each spectrum iProphet picks rank 1 hit info from either one of the three search results (the one with largest PeptideProphet probability?) to report in iProphet PepXML. Therefore for some spectrum, Leucine-version is reported, but sometimes Isoleucine-version gets reported.

I was concerned about whether the inconsistent ranking would affect iProphet scoring, but then as you mentioned, iProphet converts I/L variant and treat them as the same sequence therefore it should be fine as far as scoring. Just in the end when we read iProphet result we need to also treat I/L variants properly.  Am I understanding it correctly?

Best,

Chih-Chiang

On Tue, Jun 12, 2018 at 12:26 PM David Shteynberg <David.Shteynberg@systemsbiology.org> wrote:

Hello Chih-Chiang,

iProphet will convert all Isoleucines reported to Leucines for comparison purposes.  Have you observed an error in your analyses?

-David

On Tue, Jun 12, 2018 at 9:06 AM, Chih-Chiang Tsou <chihchi...@gmail.com> wrote:

Dear TPP developers,

I have a question regarding how iProphet handles ambiguous I/L IDs from different search engine. I am using X!Tandem, Comet, and MSGF+ and combine the results using iProphet, and noticed that the three search engines have different rules ranking I/L hits.

So how does iProphet handle such case if different search engines give different I/L variant as top ranking PSM? Does iProphet threat different I/L variant peptide as different IDs and therefore penalizing the final probability score, or iProphet considers them as the same ID?

Thanks,

Chih-Chiang

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discuss+unsubscribe@googlegroups.com.
To post to this group, send email to spctools-discuss@googlegroups.com.

Visit this group at https://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to a topic in the Google Groups "spctools-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/spctools-discuss/VDEKdnL5iEQ/unsubscribe.

To unsubscribe from this group and all its topics, send an email to spctools-discuss+unsubscribe@googlegroups.com.
To post to this group, send email to spctools-discuss@googlegroups.com.

Visit this group at https://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to the Google Groups "spctools-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discuss+unsubscribe@googlegroups.com.
To post to this group, send email to spctools-discuss@googlegroups.com.