Fragments used in quantification

[hxxhust163]

Hi

Thanks for your work in DIA-NN, it really of high performance.

I used DIANN by: diann-1.8.1 --dir /HYE124 --threads 32 --verbose 4 --out lib-base --lib lib.tsv --fasta human_yeast_ecoli_20140403_iRT_reverse.fasta --mass-acc 20 --mass-acc-ms1 10 --smart-profiling --reanalyse --matrices --qvalue 0.01 --reannotate --report-lib-info

And in the main out, I can find a column named 'Fragment.Info' showing fragments used in quantification. I wonder how these fragments are determined. Were they from the library i specified or predicted by deep learning? I checked and found that some fragments in the main output were not in my library. So I am writing for help. Thanks very much!

Showing an example:

this is the main output showing the fragments info of precursor (UniMod:5)AAAAAAGAASGLPGPVAQGLK2

b8/598.2949 and y3/317.2194 are not in my library

and this is the fragments in my library

[vdemichev]

You are using MBR and smart profiling, hence some of the spectra will be replaced with empirical ones during the second pass, I guess that's what you are observing. You can check the lib-base-lib.tsv to see what spectrum was saved there.

[hxxhust163]

Oh, I figure this out. Thanks very much!

[hxxhust163]

Hi, excuse me, I have another question.

In the main output, I find a column named 'Fragment.Quant.Raw' showing the quantity of the corresponding fragment used in quantification. And the column named 'Precursor.Quantity' showing the quantification of the precursor. My question is how the quantification of the precursor is calculated? I guess it is calculated by summing up the values in 'Fragment.Quant.Raw'. But I sum up the values to find that the sum is not equal to the value in 'Precursor.Quantity'. Obviously, my guess is wrong. Can you please help me figure it out? Thanks very much!

Here, I specify 6 fragments in my spectral library used in DIANN. Following is a subset of my main out