As per ICH Q3 the minimum area of the API related compounds MUST be no less than 0.050% w/w of the drug products 'maximum daily dosage'.  For a tablet of 10 mg taken daily this is 5 ug.

Your SIN is not using a reference standard (NIST traceable) pure enough to meet this criteria (thus all the peaks).  Try 98.0 or even 99.0% next time (particularly for MS)!

I agree with John.  In the TIC, determine the total area under all the GC peaks then calculate the area of the known peaks as a proportion of this.  In addition, I would just make the comment that you are probably setting your criteria for identification too high and are probably missing some perfectly good identifications against the NIST database.  Software identifications should be treated with some caution, eg isomers such as hexoses will have identical MS but different retention times.  Compare with standards if at all possible and manually check the comparison with the NIST library.

I would suggest you to that apart from NIST matching and authentic standards, you also need to calculate retention indices or kovats indices by running a straight chain n-alkanes and check the experimental value obatined with that of literature

I agree with Charles.  But this desired quantity request is strange to me.  The proportion of all identified peaks to all the peaks has nothing to do with the percent of all components detected or not.  Furthermore, there is no that the unidentified  peaks are due to single components.

You are right Wayne - it is an odd request and not a statistic I can recall seeing reported anywhere.  As you say it is fairly meaningless and this brings me to my point which is that you don't have to agree with unreasonable or silly requests from reviewers.  They are human and make mistakes as well. You need to politlely respond saying you don't understand the need for this metric and explain the shortcomings as Wayne has outlined.  Even reporting of the % peaks identified is not particularly useful - it is entirely a reflection of the limitations of current databases, the complexity and novelty of your particular sample and the ability of your software to deconvolute  and match the spectra from co-chromatographing peaks. To that list you can also add instrument (eg quad vs fast scanning ToF or low res vs high res) and column (bonded phase and length), derivatives etc etc 

Actually it does make sense. If I publish a chromatogram where I only run the peaks against the NIST library and report just the compounds "detected" then I have not made a legitimate attempt to identify the compounds present. What if I am only reporting identifications for 98% of my sample as "unknown"? Library matching against the NIST library is notoriously questionable. The reviewer may have noted that much of the sample went unreported.

The point I am making is that metabolite "identification" is problematic subject to limitations beyond the immediate control of the experimenter and hence an arbitrary metric such as % peak area of the TIC identified is meaningless and not necessarily a reflection of any  "legitimate attempt" at identification by the experimenter. If a sample contains a lot of analytes for which there is no MS match then that is the way it is, until databases are updated or somebody does a real job of identification.  The process of structural elucidation consists of isolating, purifying, gathering spectral data (NMR, UV/Vis, IR, high res MS) and finally making a comparison with an authentic standard.

If an authentic standard exists. 

Many unknowns are just that -- unknowns.  Any elucidation of a portion of the mixture is valuable.  Maybe one of the unknowns is more important but not identifiable.    And, it is not just metabolites.   I remember one case I had in which I "identified" an impurity  of a compound as a chemical poison slowing down the reaction without ever knowing what it was by the changing intensity of one nmr peak that did not belong to the reactant's nmr spectrum by comparing the intensity to the reported "product works" and "product does not work" for a set of different batches.  Manufacturer was able to locate a supplier of reactant material that had an extremely small amount of whatever it was.  I again agree with Charles that spectral elucidation is burdensome, but he left out chiral determination, chromatography to assure purity, and other tests possibly needed.  If the paper in question identifies important components in the mixture studied, even if it is only 2 percent of some percent of the total detected and not detected components, the requested % peak error is still, to me, worthless.  What if changing the solvent extraction process gives more components or less components detected but not identifiable, but the same peak area % of detected components?  Is anything gained from this information?  What does a lower peak area % mean compared to a higher peak area % if only the same peaks are the only ones identifiable? 

Spot on on Wayne.  Also you raise a good point about the need to carryout a comprehensive identification. In metabolomic studies, for example, where different biological states are being compared, one's interest lies in looking at what has changed.  So if the concentration of an "unknown" is essentially unaltered, it is of no particular interest and its identification is irrelevant.  However, the "unknown" that has a changed concentration is definitely of interest and its identification can then be prioritised.

Hello,

I would also translate "proportion for the peak area of identifiable compounds based on NIST database to the total detected products used GC-MS" into ratio between "sum of area counts for identified peaks divided by area counts of the TIC" . This gives you some idea about how much (as % of the sum of amounts) of the already present compounds were identified. To my opinion this is not "meaningless" as Charles has stated, but i agree it is a limited tool. The area response of the several compounds are varying much. But EI ionization in GC-MS is a more universal ionization type compared to other instrumentation, therefore the comparison has its worth. You can support this additionally with GC-FID data where the area response of hydrocarbons is very similiar. It is easy to criticize this approach, but to my knowledge there is no other practicable chromatographic approach available to estimate the extent of the identification work in a sample. If this is note the case Wayne and Charles should provide us better alternatives.

By the way, I support Asit that apart from NIST matching and authentic standards, it is recommendable to calculate retention indices. NIST supports You with respective data.

You can calculate anything possible from the data.  I am just pointing out that there is no assurance the data is a complete pro file of what is present in the sample, only of what is detectable by this method.  .  I do not think that is a meaningful calculation.  If , others do, it is up to them to explain its value.