Closed sjanssen2 closed 7 years ago
This table aggregates over all three complexity types, i.e. operates on a set of gold standard files. Is this possible with the current code? Could that be a potential usecase?
Stefan,
great to have you on board here and suggesting useful things!
best, Alice
Prof. Alice C. McHardy Computational Biology of Infection Research Helmholtz Centre for Infection Research
From: Stefan Janssen [notifications@github.com] Sent: Monday, October 30, 2017 9:06 PM To: CAMI-challenge/OPAL Cc: Subscribed Subject: Re: [CAMI-challenge/OPAL] ranking table (#13)
This table aggregates over all three complexity types, i.e. operates on a set of gold standard files. Is this possible with the current code? Could that be a potential usecase?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub<redir.aspx?REF=Z54r8_wn9jLMlQs1moiXLVfZYoYODVeiof79zRRRCWrXRrcA_SDVCAFodHRwczovL2dpdGh1Yi5jb20vQ0FNSS1jaGFsbGVuZ2UvT1BBTC9pc3N1ZXMvMTMjaXNzdWVjb21tZW50LTM0MDU2Nzk0MA..>, or mute the thread<redir.aspx?REF=m_7loUlu-nL9CyoVwhRxlVS-UHqnyei_oOUcNYcl_AbXRrcA_SDVCAFodHRwczovL2dpdGh1Yi5jb20vbm90aWZpY2F0aW9ucy91bnN1YnNjcmliZS1hdXRoL0FDQUg0V3FCYzBMVk5GdG1sSW5BUE1hWkRvS0p0cS1Qa3M1c3hpdy1nYUpwWk00UUx4YUo.>.
Helmholtz-Zentrum für Infektionsforschung GmbH | Inhoffenstraße 7 | 38124 Braunschweig | www.helmholtz-hzi.de
Vorsitzende des Aufsichtsrates: MinDir’in Bärbel Brumme-Bothe, Bundesministerium für Bildung und Forschung Stellvertreter: MinDirig Rüdiger Eichel, Niedersächsisches Ministerium für Wissenschaft und Kultur Geschäftsführung: Prof. Dr. Dirk Heinz; Silke Tannapfel Gesellschaft mit beschränkter Haftung (GmbH) Sitz der Gesellschaft: Braunschweig Handelsregister: Amtsgericht Braunschweig, HRB 477
my pleasure :-) Thank you.
@sjanssen2 Even if only a single sample/complexity type, this could still be of use since it's also aggregating over taxonomic level (so gives an impression of "overall performance" over taxonomic ranks).
Already addressed in #1
I think the ranking table (Figure 3 c of CAMI publication) is really useful to get an overall impression of how well a profiler performs. Are you going to integrate this table into the code base?