RNA-Seq. An alternative to microarray. Grow cells or isolate tissue (brain, liver, muscle) Isolate total RNA Isolate mRNA from total RNA (poly A+ select) Fragment RNA Make and amplify cDNA Sequence the ends of the cDNA Map the sequences to the human genome
An alternative to microarray
Isolate total RNA
Isolate mRNA from total RNA (poly A+ select)
Make and amplify cDNA
Sequence the ends of the cDNA
Map the sequences to the human genome
Count the number of sequence tags at each known gene (and at locations for which no gene is known)
Correct for background
Analyze the dataSteps
1) Load Samples to Flow Cell
2. Attach DNA to Surface
3. Bridge Amplification
8 Lanes are loaded onto the flow cell for simultaneous analysis
Single stranded DNA fragments bind randomly to the inside surface of the flow cell.
Unlabeled nucleotides and enzyme are added to initiate solid-phase bridge amplification.
4. Fragments Become
5. Double Stranded Molecules
6. Amplification is Completed
The enzyme incorporates nucleotides to build double stranded bridges on the solid-phase substrate.
Denaturation leaves single-stranded template anchored to the substrate
Several million dense clusters of double stranded DNA are generated in each channel of the flow cell.
1. Determine 1st Base
2. Image 1st Base
3. Determine 2nd Base
The first sequencing cycle is initiated by adding all 4 labeled reversible terminators, primers, and DNA polymerase to the flow cell
After laser excitation, an image of the emitted fluorescence from each cluster on the flow cell is captured.
The 2nd sequencing cycle is initiated by adding all 4 labeled reversible terminators and enzymes.
4. Image 2nd Base
5. Sequence Read Continues Over Multiple Chemistry Cycles
6. Align and Map Data
After laser excitation, image data is collected like before. The identity of the 2nd base for each cluster is recorded.
35 cycles of sequencing are repeated to determine the sequence of bases in a given fragment a single base at a time.
Align data and map the sequences to the reference genome.
What can you learn from one but not from the other?
How is the primary data acquired?
How are systematic biases eliminated?
How do you normalize
How would you look for differential expression?
How would you cluster?
How can you combine data from multiple experiments?
Which is more sensitive?
What kinds of additional software do you need?What are the similarities and differences?