We received an email from Arima support with suggested changes specifically for Arima Hi-C. Arima is a known kit provider for Hi-C library preparation using restriction enzymes for fragmentation.
Even if it were to proof difficult to implement the changes in a generalised Hi-C pipeline, the input can be very valuable to different pipeline users and it would be good if the information on how to adapt the pipeline to Arima Hi-C libraries would not be lost.
Here is the email from Arima sevice:
Dear NF-core Hi-C Development Team,
My name is Xiang Zhou, and I am a staff bioinformatics scientist at Arima Genomics. I have been evaluating the NF-core/hic pipeline (https://nf-co.re/hic/2.1.0/) for the analysis of Hi-C data generated using Arima’s kits. I appreciate the comprehensive and well-documented nature of the pipeline, as well as its utility to the broader genomics community.
During our internal benchmarking, we noticed that the current version of the pipeline uses HiC-Pro for read mapping but does not include a trimming step to remove the first 5 nucleotides from the 5’-end of both R1 and R2 reads. This trimming is a critical preprocessing step for Arima Hi-C data.
As described in our published user guide (Arima Mapping User Guide), trimming the first 5 bases improves data quality by removing an over-represented 3 bp molecular barcode (UMI) along with 2 dark bases. This modification has consistently resulted in improved mapping performance and downstream analysis in our internal assessments.
Additionally, in order to prevent unnecessary data loss, the following modifications must also be made to the HiC-Pro configuration file (assuming 2-enzyme chemistry):
GENOME_FRAGMENT = XXXXXX_GATC_GANTC.bed
LIGATION_SITE = GATCGATC,GANTGATC,GANTANTC,GATCANTC
MIN_FRAG_SIZE = 10
MAX_FRAG_SIZE = 100000
MIN_INSERT_SIZE = 100
MAX_INSERT_SIZE = 1000
We recommend that these steps be integrated into the preprocessing workflow of the NF-core/hic pipeline, particularly when users specify that the data originate from Arima Genomics kits. The addition of a conditional trimming step and the modification to the configuration file would help ensure more accurate alignment and better-quality contact maps for users working with Arima-generated data.
Please let us know if we can provide additional details, support, or test datasets to assist with this implementation.
Thank you for your continued work in building high-quality, community-driven bioinformatics pipelines.
Sincerely,
Arima Genomics Technical Support Team
We received an email from Arima support with suggested changes specifically for Arima Hi-C. Arima is a known kit provider for Hi-C library preparation using restriction enzymes for fragmentation.
Even if it were to proof difficult to implement the changes in a generalised Hi-C pipeline, the input can be very valuable to different pipeline users and it would be good if the information on how to adapt the pipeline to Arima Hi-C libraries would not be lost.
Here is the email from Arima sevice:
Dear NF-core Hi-C Development Team,
My name is Xiang Zhou, and I am a staff bioinformatics scientist at Arima Genomics. I have been evaluating the NF-core/hic pipeline (https://nf-co.re/hic/2.1.0/) for the analysis of Hi-C data generated using Arima’s kits. I appreciate the comprehensive and well-documented nature of the pipeline, as well as its utility to the broader genomics community.
During our internal benchmarking, we noticed that the current version of the pipeline uses HiC-Pro for read mapping but does not include a trimming step to remove the first 5 nucleotides from the 5’-end of both R1 and R2 reads. This trimming is a critical preprocessing step for Arima Hi-C data.
As described in our published user guide (Arima Mapping User Guide), trimming the first 5 bases improves data quality by removing an over-represented 3 bp molecular barcode (UMI) along with 2 dark bases. This modification has consistently resulted in improved mapping performance and downstream analysis in our internal assessments.
Additionally, in order to prevent unnecessary data loss, the following modifications must also be made to the HiC-Pro configuration file (assuming 2-enzyme chemistry):
We recommend that these steps be integrated into the preprocessing workflow of the NF-core/hic pipeline, particularly when users specify that the data originate from Arima Genomics kits. The addition of a conditional trimming step and the modification to the configuration file would help ensure more accurate alignment and better-quality contact maps for users working with Arima-generated data.
Please let us know if we can provide additional details, support, or test datasets to assist with this implementation.
Thank you for your continued work in building high-quality, community-driven bioinformatics pipelines.
Sincerely,
Arima Genomics Technical Support Team