Hi Kawitha,

the RIN is a function of the degradation of the two most prominent peaks in RNA preps, which are the ribosomal peaks. If those cannot be detected, that indicates strong degradation of your samples. The concentration you described sounds good, more is unlikely to help. Just a gut feeling, but most likely not the issue (since you sent before with good results): you did not enrich for polyA-tails before sending the samples, did you? If you get rid of the ribosomal RNA, that would be an issue. Maybe run an RNA-gel after purification to get a better idea of what your samples look like.

Cheers,

Lukas

• I feel the question does not provide complete information if you depleted rRNA as also speculated in post above. If they are rRNA depleted samples, RIN value does not applies to such samples however, the samples can still be used for sequencing or any other application.
• When you can the 260/280 ratio, most probably you are using nanodrop for quantification. I would recommend to use qubit or bioanalyzer for QC of samples if the samples are going to be used for sensitive analysis. Nanodrop reading can be sometimes misleading and therefore not recommended by most of the sequencing/analysis facilities.
• "How to fix RIN value"?

>> I think it is nothing which can be fixed. You are analysing the final eluent which is already processed and quality cannot be fixed. However, you can do the whole process again with more care and see what is the final quality.

Don't use the samples with the low RIN. The community standard for RIN is greater than 7.0. It is possible to make stranded libraries which would allow you to use fragmented RNA. Basically, the issue is how the get rid of contaminating RNA, like ribosomal RNA. The standard method is poly A selection, this will not work well enough if the RNA is fragmented. There are antisense oligos bound to magnetic beads that can be used to deplete ribosomal RNA for stranded library production which depends on fragmented RNA, but it doesn't work as well and can introduce artifacts, similar to a batch effect in your downstream data analysis. It is important to do PCA analysis with your replicates to ensure that there isn't the same magnitude of difference between groups as replicates. This is why in a relative comparison analysis you would like to have as much precision as possible.

If your cells are dying that can cause the RNA to be fragmented. You may want to keep the cells on ice and minimize the manipulation if possible. You can look at them with Trypan blue exclusion, but sometimes this is misleading. Annexin V staining is more authoritative, this can help you optimize the tissue extraction and cell isolation. Red cell lysis buffer is really hard on cells, you could try to do density centrifugation through percoll. The issue is that if your cells are mostly dead it wont really affect the OD ratio and your extraction could be perfectly fine. I recommend that you separate an aliquot of your total RNA for measurement of the RIN and concentration by the company making the libraries. I use 2ul in a screw-top tube with an O-ring. Often a core will expose your sample to freeze/thaw for measurements before making libraries this jack the concentration as RNA degrades from freeze/thaw. If after all this, you still can't get good RIN, you may need to use a hybrid polyA stranded kit. KAPA makes one for low input. Try to get matched concentrations and increase the number of biological replicates, then you can drop them after PCA if necessary. You can't do anything about accuracy, but precision is all you. Good luck!

One last comment. I though you were doing bulk RNA-seq. Sorry. I wouldnt use nanostring unless someone gave them to me for free. It is old technology, like microarrays. There is a GeoMX from that company, but its spatial. If you find out who your rep is for Novogene, bulk RNA-seq is probably very similar in cost.