Repcred testing - performance on IGH

[bcorrie]

In order to find a range of repertoires from a range of studies for testing, I chose a PCR target of IGH and then went through a range of data set sizes from 100 - 2,000,000. This managed to get repertoires from 7 different studies which is good coverage.

• 110 sequences, PRJNA206548,LN10,LN10,axillarylymphnode,IGH, Success
• 1,001 sequences, SRP001460,CLLA,CLLA-e,lymphnode,IGH, Success
• 9,933 sequences, PRJEB1289,CLL_10_3-sc-2013-01-11T14:16:25Z-1548353,CLL_10_3,blood,IGH, Success
• 99,656 sequences, IR-Binder-000001,Pt-20,Pt-20-1,blood,IGH, Success
• 1,014,756 sequences, PRJNA628125,7452,M369-S003,venousblood,IGH, Success
• 2,017,519 sequences, PRJNA280743,SLE5-262,SLE5-262,blood,IGH, Success

All tests succeeded and generated what look to me to be like reasonable reports.

Congrats! 8-)

[bcorrie]

I am currently running a 4M sequence test, a repertoire of this size failed with out of memory in #37, I am going to see if that occurs again.

[bcorrie]

Yes, the 4M sequence job did indeed run out of memory with a memory allocation of 7.8 GB.

Sample ID: F1NCpre PCR target: IGH

[bcorrie]

I am re-running this with 15GB to see if it runs, will let you know...

[bcorrie]

You can probably close this issue, since the memory issue is captured in #37