Focusing

1. open "Manual Control/Setup" window;
2. fill the FC with scanning mix: pump: [to: flowcell · solution: 3 · volume: 100 · asp. rate: 100 · disp. rate: 2500] ⇒ Enter;
3. move the stage:
   • move: [Z: 0] ⇒ Enter;
   • move: [X: 0 · Y: 0] ⇒ Enter;
     moves to the left edge of Lane 1
4. initial focusing:
   • set up: [Laser: Green · Filter: T · Exposure: 100msec] ⇒ Take picture;
   • right-click over the image — Auto Scale/On;
   • zoom in to approximately 1/10 the image;
   • find focus (Z) with the accuracy about 1000 (steps max 2500nm) (check the left edge of lane 1 is visible as a sharp line);
   • check there are clusters;
5. setting X:
   • bring the left edge of lane 1 to the crosshair;
     increase X value to move the edge to the right; decrease — to the left
     width of the tile is 760µm or 2048px: 0.371µm/px;
     pop-up window shows cursor position in pixels.
   • set up: [Laser: Green · Filter: None · Exposure: 3msec] ⇒ Take picture;
     the edge of Lane 1 should appear as a blurred wide line. The crosshair should be within 3-5 pixels of the right edge
   • "Instrument" ⇒ "Set Coordinate System" ⇒ "Set Current X as Origin" ⇒ OK;
6. setting XY tilt:
   • move: [Z: 0] ⇒ Enter;
   • move: [X: 0 · Y: 35000] ⇒ Enter
     moves to the upper-left edge of Lane 1
   • move the stage in the X axis to bring the edge of Lane 1 to the crosshair;
   • "Instrument" ⇒ "Set Coordinate System" ⇒ "Set Current X as top-left edge to determine XY drift" ⇒ OK;
7. confirming the XY-setting:
   • move: [Lane: 4 · Col: 1 · Row: 25] ⇒ Enter;
   • set up: [Laser: Green · Filter: T · Exposure: 100msec] ⇒ Take picture;
     adjust focus if necessary: should be a thin black band on the left side of the image
   • move: [Lane: 4 · Col: 2 · Row: 25] ⇒ Enter;
   • take a picture
     should be a thin black band on the right side of the image
8. setting Z (roughly):
   • move: [Lane: 4 · Col: 1 · Row: 25] ⇒ Enter;
   • set up: [Laser: Green · Filter: T · Exposure: 100msec] ⇒ Take picture;
   • find focus (Z) with the accuracy about 1000 nm (initial steps: 2500nm)
     zoom in (~10x) to see clusters;
     move the cursor on the image to see FQ value. It should be as high as possible: ~60-80;
   • set up: [Laser: Red · Filter: A· Exposure: 100msec] ⇒ Take picture
   • using the FQ value find focus (Z) with the accuracy about 100 nm;
   • "Instrument" ⇒ "Set Coordinate System" ⇒ "Set Current Z as Origin" ⇒ OK
9. setting Z (accurately):

   manual focusing move Z from "-500nm" to "+500nm" in 100nm steps; take pictures in every step and record FQ values for channels A, C, G, T; select the Z value for which the FQ sum for the four channels is the highest; "Instrument" ⇒ "Set Coordinate System" ⇒ "Set Current Z as Origin" ⇒ OK

   automatic focusing procedure is very formal and have no internal logic. It is better to use original step-by-step description Setting up focus on a GA2 using FocalAssist Utility.pdf.

Checking quality

1. open "Run" window;
2. autofocus calibration:
   • resume the 1^st base incorporation recipe from the "User wait" step before "Incorporation";
   • click "Yes" for autofocus calibration;
   • check the calibration values (goodness of fit ≥0,99; sensitivity: 350-400nM) ⇒ Accept;
   • press OK to start imaging;
3. run Browser data:
   • look through taken images;
   • wait till the Run Browser report appears:
     * write number of clusters (should be 90-120x10³ per tile);
     * FC tilt: difference between min and max focus stage values should be less than 15000 nm;
     * intensity values:

     Confidence A & C G T High > 650 > 1000 > 1200 Reasonable > 350 > 650 > 700 Low < 250 < 350 < 400

Notes

Genome Analyses uses a laser beam as a light source. It come through (i) prism, (ii) mineral oil, (iii) FlowCell basement and is internally reflected from the bottom of the channel. In theory, the laser beam should be reflected completely on the "glass-water" surface and the picture should be completely dark even without filter. On practice, all tiny scratches of this surface shine brightly ("glass-oil" scratches are invisible, because glass and oil have the same refraction coefficient). These bright scratches may be used for the preliminary focusing.

The diameter of the laser beam is about the same size as a size of a picture, so:

• purity and quality of the optical surfaces should be extremely high;
• if laser beam is occassionally shifted, it would be impossible to set it back w/o special equipment.





Fig. 2. Focusing. a. Z=250000; b. Z=150000; c. Z=50000; d. Z=0 (focus); e. clusters from the previous run.





Fig. 3. Position of the objective. a. Z=9000, objective is too high: circles instead of dots; b. Z=focus; c. Z=-10000, objective is too low: blurring, weakening.





Fig. 4. Clusters. Clusters at about optimal density.