Seeking Insights on Error Discrepancies in Combined ONT and HiFi Sequencing Data”

[Bank-tidy]

Seeking Insights on Error Discrepancies in Combined ONT and HiFi Sequencing Data”

[JiaoLaboratory]

hello,
"the third-generation data in the image exhibits fewer errors compared to the second-generation data, which shows more errors." whic is not the case.

by default, CRAQ would report two stypes of assembly error CRE(like small local indels,which could be spanned by SMS reads) and CSE (like structral misjunction errors, which could not be spanned by SMS reads). Both CRE ans CSE are detected by combinning SMS and NGS data (not using NGS data to detecte CRE, or using SMS data to detected CSE ,separately). for your result, we can conclude the number of CREs over CSEs, that is normal.

for CRE (CRAQ cound not perform error correction, because such errors in fact cound generally corrected via multiple rounds of polish,like using polin, Racon, et al.) while, for CSE, which implies a structural mis-join that affects the overall assembly continuity, and CRAQ will separate the contig by interrupting the contig at CSE location for futher scaffolding by using hic or bionano maps.

from the circos, I can conclude only one CSE be found at chr9, (must be a high S-AQI score, the result means an high continuity for your assembly). While I can see also some CREs at each chromosome, means still some local indel-errors within your assembly. If you want futher quality improvement, maybe some rounds of polish useful. In fact, I donot know the exact counts of CREs, if the R-AQI score is over 95(reference huality), I think it is really good quality, which depends on you to polish or not.

[Bank-tidy]

Thank you so much for your valuable insights and guidance on our genome assembly analysis using CRAQ!