Sample submission guideline

NOTE: REFER TO (https://docs.google.com/spreadsheets/d/12ghgV_agYnepJ4B8aMvt7RLX8n6RFVRz4j3CEHhFIAU/edit?usp=sharing)[https://docs.google.com/spreadsheets/d/12ghgV_agYnepJ4B8aMvt7RLX8n6RFVRz4j3CEHhFIAU/edit?usp=sharing] INSTEAD FOR SAMPLE SHEET SUBMISSION

When submitting data to the lab, please fill out the data submission file, rename it to properly reflect the project, and send it to clauwaer@umich.edu.

Some notes:

• When using Excel, please save as a .csv file format after editing. The file is parsed by our pipelines and therefore needs to be in a flat file format.
• Refrain from using comma’s within field entries.
• Use the underscore character ('_') instead of spaces in the main Table (not the Metadata section).

When renaming the sample submission sheet, follow the naming convention YYYYMMDD_<project_id>_sample_sheet.csv. Where the project ID reflects the experimental setup of the data (e.g., 20240615_MS_MBL_MYC_groups_sample_sheet.csv, 20220927_A673_bromo_menin_inhibitor_treatment_sample_sheet.csv.

This documentation provides additional information to fill out every field.

Metadata

• PI name: Name of Principal Investigator of the lab
• Date: Date at which samples were created. Use YYYYMMDD format (don't use additional '/' or '-').
• Submitter name: Your name
• Submitter email: Your institution email
• Project title: Short title used as reference of the sample study
• Project description: Short description of the purpose of the study. Make sure to use a single field within the csv file.

Table

Please fill in NA if field is not applicable. Additional columns can be added when missing.

• run_id: File name of the data generated from the sample. Include the file extension. Leave empty or use sample reference if filling out before data is processed

• smart_id: ID used for downstream processing of the samples. Make sure this value is intelligible and differentiates samples from each other that are not supposed to be merged, as this smart_id is used in all the report files. Use properties essential to the experimental set-up (E.g., KO vs. Control, Dose rates, replicate number). An example for RNA_seq data featuring two biological replicates:

  run_id	smart_id	data_type	read_end	replicate_num	treatment_id
  123_0313_R1.fastq.gz	DIPG13_K27M_5Gy_1	RNA_seq	R1	1	5Gy
  123_0313_R2.fastq.gz	DIPG13_K27M_5Gy_1	RNA_seq	R2	1	5Gy
  123_0314_R1.fastq.gz	DIPG13_K27M_5Gy_2	RNA_seq	R1	2	5Gy
  123_0314_R2.fastq.gz	DIPG13_K27M_5Gy_2	RNA_seq	R2	2	5Gy
  123_0315_R1.fastq.gz	DIPG13_K27M_10Gy_1	RNA_seq	R1	1	10Gy
  123_0315_R2.fastq.gz	DIPG13_K27M_10Gy_1	RNA_seq	R2	1	10Gy
  123_0316_R1.fastq.gz	DIPG13_K27M_10Gy_2	RNA_seq	R1	2	10Gy
  123_0316_R2.fastq.gz	DIPG13_K27M_10Gy_2	RNA_seq	R2	2	10Gy

  This smart_id links samples through several rules. Samples are supposed to share the same smart_id when:

  • They are paired-end samples (e.g. R1 and R2; see field read_end and above example)
  • They only differ from their data protocol (e.g. RNA_seq/Ribo_seq; see field data_type)
  • They are meant to be merged (reads are combined), e.g. multiple technical replicates run for greater read depth. Don't use the same smart_id for biological replicates!

  Samples are only merged if they have identical smart_ids and also have the same values in the data_type and read_end field.

• data_type: Please use one of the following values:

  • RNA_seq
  • Ribo_seq
  • ONTRNA_seq
  • Disome_seq
  • ChIP_seq
  • Exome_seq
  • WGS
  • Other (detail further in description)

• species: Species identifier

• sample_type: Please use one of the following: autopsy/tissue_sample/cell_line

• source_id: Cell line code, patient ID or other unique identifiers.

• is_paired_end: Please use one of the following: true/false

• read_end: Please use on of the following: R1/R2/NA

• experiment_type: Could be one of the following: overexpression/knockdown/drug_treatment.

• test_or_control: Used by the differential expression pipeline to understand which sample is the reference. Use either test or control.

• location_id

• disease_id

• treatment_id: Use a short string to denote one or multiple treatment groups, such as the amount of administered drug and/or the gene that is knocked out/overexpressed (e.g. K27M_noIR, K27M_5Gy, K27M_10Gy, KO_noIR, KO_5Gy, KO_10Gy). In other cases, for large sample cohorts, it might be more applicable to use sample types (e.g., Cancer type MBL_group_3, MBL_group_2, etc.). Treatment ID's determine how samples are grouped for differential expression analysis, where each group of samples is compared against all other groups. Replicates should be grouped together (i.e. sharing the same treatment ID).

• replicate_num: Biological replicate number (integer)

• sequencing_instrument: sequencing instrument on which the samples were run on

• batch_date: date of processing. Use YYYYMMDD format (don't use additional '/' or '-'). Can be different when project data is retrieved over multiple batches. Or when Ribo-seq and RNA-seq are processed at different dates.

• description: Free from description of the sample. Can be used for more comments on the treatment etc.

• Custom columns can be added for referencing information that is relevant.