For the application being discussed you are correct and I do feel better when there are 1000 events to study. However, in studying lymphoid tumors using a multiparameter approach we can make an accurate diagnosis with 100 cells- no problem. If a patient with hairy cell leukemia, with surface lambda light chin expression, bright CD22, bright CD11c, and negative kappa, if you identify 100 cells with this immunophenotype I all it diagnostic. You just can't pick a number for all applications. Maryalice > >I do human Monocytes all day here, In most of my studies a 100ul wb blood >sample can easily yield 1000 mono's in about 7-10 min from a lysed sample >using a FSC trigger. >for a whole blood sample ( no lyse) using a flour trigger instead of the >FSC and dilute 1/20 (to reduce coincidence) and you can get 1000 in about 10 >min on LOW flow rate.... > >A 100 event file will contain about 2% noise ( more if whole blood)..this >drops after 300 events...statistically around 1000 is a safe bet. > >-----Original Message----- >From: bunny [mailto:bunny@cotleur.com] >Sent: Monday, April 10, 2000 5:51 PM >To: cyto-inbox >Subject: analysis of few cells > > > >Fellow Flowers: >I wonder if I could gather some opinions on releveance as it relates to >minimum numbers of events collected. >I am assisting a lab in developing FACS analysis of PBMC's and monocytes >in CSF. They informed me that when they run FACS, they are "lucky" to >get 100 events. I raised the questions of relevance, stating that those >100 "events" could as easily contain some non-cell events as cell >events. They insist their data is good becasue it compares somewhat (?) >with a lab in Europe they are collaborating with. We have had many >conversations on this. >(I should mention- neither lab has direct FACS experience. They got all >their protocols second hand, and just collect what they collect). >Additional information: they are looking at chemokine expression >(another difficult task) on these few cells- and I believe they are >using the PBMC's to set up isotype controls. I doubt they even use comp >controls. > >Please throw your 2cents my way. And if any of you do this type of >analysis (few cells) with alternative techniques- I'd love to hear about it! >Thanks- >-- > > > >Bunny > > > > > > >******************************************************* >Bunny Cotleur +*+ Bunny Cotleur >Cleveland Clinic Foundation *+* 2001 Lester RD >Neurosciences NC30 +*+ Valley City, OH 44280 >9500 Euclid Avenue *+* 330-483-4800 >Cleveland, OH 44195 +*+ bunny@cotleur.com >216-444-1164 *+* >cotleua@ccf.org +*+ > >******************************************************* >When you do something, you should burn yourself completely, like a good >bonfire, leaving no trace of yourself. >(Shunryu Suzuki) Maryalice Stetler-Stevenson Director Flow Cytometry Unit Laboratory of Pathology, NCI, NIH
This archive was generated by hypermail 2b29 : Sat Mar 10 2001 - 19:31:16 EST