Chromatography Forum

I am fairly new to GCMS data analysis, so I apologize if this question doesn't make sense. I am running MSD Chemstation G1701EA E.02.02.1431 on a windows 11 laptop and I'm trying to add some more libraries to aid with identifying lipids from sediment samples as part of my PhD thesis.

I was given the NIST 2008 library to start off with, but I'd also like to add some more recent libraries such as the LipidBlast library found here . Following some advice from the forum, I used LIB2NIST to convert a .msp library to a jcamp (.hpj) library, then used the parametric retrieval option on chemstation to read the jcamp file into a .L library. This reads entries one at a time into a new .L library folder (~20/sec), so would take ~7hrs to read a library with 500k spectra.

The issue is that after reading a couple thousand spectra (usually around 25k but sometimes earlier) the program crashes and sometimes returns the error, "Sorry, a system error has occurred! Program will close now. Examine LOGGING file if LOGGING is on."

I tried searching through the logging file (pretty sure its daorphstart1, but not entirely sure) but I'm not sure what to look for.

I'm wondering if anyone here knows what may be the source of my problem, ideas on how to find the problem, or if there are easier ways to convert .msp/.sdf libraries to .L libraries that don't involve expensive software.

Thanks!

As far I know - jcamp are text files.
Split your jcamp file into smaller ones.

import them one-by one. One of them fails, delete the broken spectrum and try again.

Ask somebody to help you with Python

As far I know - jcamp are text files.
Split your jcamp file into smaller ones.

import them one-by one. One of them fails, delete the broken spectrum and try again.

Ask somebody to help you with Python

Thanks for the reply, I was able to write a script (thank you chatgpt ) that split the original ~500k .HPJ into 25 batches. Reading them one by one as .L libraries isn't a problem, but I haven't been able to find a way to merge them together.

Another potential solution I thought of was to instead append a single .L library multiple times and then noting at which entry chemstation crashed. Then, I "shaved" a copy of my original .HPJ library to start at the entry where I crashed.

So far, I've been able to get the appended .L library to ~90K entries over seven attempts, but with each appending cycle the rate of entry reading takes longer and crashes at 4-8k reads instead of ~20k.

Until I find a way to merge multiple .L libraries together, I am planning on following my appending approach for as long as possible.