View source for BioMicroCenter:FAQ

{{BioMicroCenter}}

== CONTACT ==
'''STOP BY:''' The BioMicro Center is located at [http://whereis.mit.edu/?go=68 68-322 on the MIT campus]. <BR>
'''EMAIL:''' The BioMicro [[BioMicroCenter:People|staff]] can be emailed at biomicro@mit.edu <BR>
'''PHONE:''' The main lab number is 617-715-4533 <BR>
Our director, [[User:Stuart S. Levine|Stuart Levine]], can be reached at 617-452-2949.

== DEFINITIONS ==
=== CORE LABS ===
The BioMicro Center is NOT an MIT institutional core, but a departmental resource. Departments that fund the operation of the BioMicro Center are given priority for all resources within the Center. Several labs that have provided major equipment donations to the BioMicro Center are also considered to be CORE. The BioMicro Center is funded by:
* The Department of Biology 
* The Department of BioEngineering
* The David H. Koch Institute for Integrative Cancer Research
* MIT Center for Environmental Health Sciences

=== ASSISTED versus WALKUP SERVICES ===
Services available in the BioMicro Center are offered either as assisted or as walk-up. Walk-up Services are those where the Center provides training and maintenance of equipment and may also provide some consumables. Scheduling for Walk-up services is available through the calendar functions on ilabs and are almost exclusively limited to MIT users except in rare cases. Non-MIT users should contact biomicro@mit.edu to schedule walk-up equipment.<BR><BR>
Assisted services are those where samples are delivered to the BioMicro Center and are analyzed by Center staff. These are set up using the [[BioMicroCenter:Forms|FORMS]] on this site or the "Request Services" tab in [https://mit.ilabsolutions.com/sc/3381/ki-genomics-core-mit-biomicro-center/?tab=about ilabs]. Assisted services cannot typically be "scheduled" with rare exceptions.

=== STANDARD versus HIGH THROUGHPUT LIBRARY PREPARATION ===
Standard library preparation is designed to maximize successful production of libraries and is used for routine work and should also be used for all precious samples. Standard library preparation includes initial sample quality control as well as at least one repeat for each library preparation if required due to sample failure. Standard library preparation is charged on a per sample basis. Standard libraries may be prepared by hand or using automation depending on throughput of the core. <BR><BR>

High Throughput library (HTL) preparation is designed to minimize cost and has many differences from standard library preparation as a result. HTLs costs do NOT include initial sample quality control nor repeats of failures of individual samples. As such, we very strongly discourage HTL library prep for precious samples or experiments with limited replicates (N<=3). Replicate counts of 4 are considered the minimum with 6 being recommended where possible. Some methods, such as HT3DGE, can never repeat failed samples, while others can be reproduced by hand, though that does create a technical variable that can impact analysis. HTLs are almost always produced using liquid handlers, typically using miniaturization, and may require submission of the samples in specific formats (noted on each library prep type). HTLs are spot checked for quality after library production. Full quality control before and after experiments as well as sample arraying and sample cleaning is available at an additional charge.<BR><BR>

We have published a manuscript highlighting some of the methods on [https://www.ncbi.nlm.nih.gov/pmc/articles/pmid/33100919/ JBT].<BR><BR>

A webinar by our director for SPT Labtech [https://www.sptlabtech.com/resources/automated-low-volume-liquid-handling-for-cost-effective-ngs-library-preparation-and-single-cell-genomics-webinar-sign-up can be found here]

== ILABS ==
The BioMicro Center uses [https://mit.ilabsolutions.com/sc/3381/ki-genomics-core-mit-biomicro-center/?tab=about ilabs] for all internal submissions and for scheduling. All users must use ilabs IF:
* They are from MIT (including Whitehead Institute or Broad Institute with MIT appointments)
* They are using an MIT cost object to pay for the services
Users not from MIT should never use ilabs for submission to the BioMicro Center <BR>
Users from labs new to MIT may have a brief grace period to not use ilabs while their labs are set up <BR> 
Users paying with POs may either use paper forms or ilabs but MUST bring the PO number with submission. We will use ilabs to import the project into our LIMS and close the project on ilabs without billing it.

=== APPROVAL ===
'''IMPORTANT''': The BioMicro Center prioritizes timely delivery of genomic data. As such, WE DO NOT HOLD PROJECTS WAITING FOR FINANCIAL APPROVAL!. If a project is not '''rejected''' by the submitting lab in a timely manner, we will proceed with the project and recharge normally.

=== COST OBJECTS ===
The BioMicro Center does not control which cost objects are available to you for your project. If no cost object is available, you will NOT be able to submit via ilabs. Cost objects are assigned to you in ilabs by your PI or a designated administrator. Having permission to spend on a cost object through MIT is NOT automatically passed to ilabs and ilabs must be set up separately. 

== DOWNLOADING DATA ==
You will be notified by email that your data is ready. The data will be placed on one of our servers for you the download. All data is typically retained for 60 days. We STRONGLY encourage you to keep a local copy of your data as soon as possible. Contact [mailto:bmc-data@mit.edu bmc-data@mit.edu] if you have any issues.

=== MIT / MIT-CERTIFICATE + VPN ===
{| class='wikitable'
 |'''LURIA'''||
* samples can be directly copied from our share /net/bmc-pub17/data/bmc/public/[LAB]/[PROJECT]
* Please ensure you are running cp from a COMPUTING NODE of Luria instead of the head node
 |-
 |'''Windows/Mac SFTP client||
* Mac: Fetch | Windows: SecureCRT/FX. Download from [https://ist.mit.edu/software/filetransfer Secure File Transfer at MIT]
 |-
 |'''Web Download <p> (MIT certificate required)'''||
https://bmc-data.mit.edu/[LAB]/[PROJECT]
|}

=== NON-MIT ===
Non-MIT users should use a SFTP client to connect to bmc-150.mit.edu with credentials emailed to you. SSH or SCP is not supported.
{| class='wikitable'
  |'''WINDOWS/MAC'''||
* SFTP client recommended: [https://filezilla-project.org Filezilla]
* A screenshot of FileZilla [[Image:FileZillaScreenshot.jpg|600px]]
  |-
  |'''UNIX / LINUX'''||
* Samples can be copied from our servers using ''sftp'' and the credentials provided.
|}
Please contact [mailto:bmc-data@mit.edu bmc-data@mit.edu] if you have difficulty obtaining your data.

=== Read archival ===

Users are strongly encouraged to retain a local copy of their data as early as possible after sequencing result delivery. 

<b>After 60 days, fastq files will be compressed on the server. Bam files, containing reads aligned either to the genome mentioned in the sample submission form or phiX (a phage genome), will be deleted to save storage space.</b><BR><BR>

Older files: Prior to 2020, bam files were maintained as archival files. Fastq files can be regenerated from the bam files (samtools sort -n to sort bam files by read names, and converted back to fastq files using bedtools bamToFastq).

== UPLOADING DATA ==

To upload your data, you can access our file server named bmc-opendata.mit.edu via the sftp protocol. 

* Request SFTP account with projects and lab information for storage setup
* Use a SFTP client (such as WinSCP, FileZilla, PuTTY and Cyberduck for Windows/Mac or sftp command for Linux)

== NON MIT USERS ==
=== Do you take samples from outside MIT? ===
The BioMicro Center is built to serve the MIT community. As such, members of the MIT community have priority on all of our services. However, if we have extra capacity, we are happy to make it available to scientists not affiliated with MIT, with the caveats that MIT samples *always* have priority and that access is finite. ''The BioMicro Center retains the right to refuse any sample.''

=== How can we ship samples to you? ===
Please email biomicro@mit.edu to arrange a drop off date and time. DNA samples should be shipped at 4C and RNA samples should be shipped on dry ice.Samples should be submitted with a [[BioMicroCenter:Forms|completed order form]] and shipped by overnight delivery to:

  MIT BioMicro Center
  31 Ames Street, Building 68-322
  Cambridge, MA 02139

An electronic copy of the full PO should be emailed to biomicro@mit.edu

=== The pricing form says "NA". What does that mean? ===
Some of our services are specifically restricted to the MIT community and we cannot offer them to outside users. Others are restricted to academic labs. Please email us at biomicro@mit.edu if you have any questions.

=== What is your billing address? ===
Payments should be made to our finance office at:

  Massachusetts Institute of Technology
  Biology Finance Office, Attn: Alison Salie
  77 Massachusetts Avenue, 68-157
  Cambridge, MA 02139

=== MIT W9 ===
[http://vpf.mit.edu/sites/default/files/static/pdf/MIT_W9_Form.pdf LINK]


<!---
== SUBMISSIONS ==
=== Volume/Concentration ===
The volume and concentration needed for submission will depend on the type of sample prep.  However, if you are submitting prepped, sequence-ready samples, we require a minimum of 10uL at a concentration of at least 2.0 nM.  This allows us to perform quality control and have enough of the library for clustering.  Any samples less than 2.0 nM cause significant difficulties with clustering the flowcell, which is why we cannot guarantee optimal read count and quality should you like us to proceed with sequencing a low concentration sample.

=== Custom Primers ===
When submitting [[BioMicroCenter:Sequencing|'''custom primers''']], be sure to verify their compatibility with the standard Illumina Primers used.  If you'd like a second opinion on the compatibility of your primers, we'd be more than happy to sit down with you and verify the primer design based on your constructs.

Another important aspect to consider is the fact that clustering conditions differ when comparing the HiSeq and NextSeq to the MiSeq.  The MiSeq clusters at about 5 degrees higher than either the NextSeq or the cBot, which is used to prepare flowcells for the HiSeq.  This means that all custom primers must be designed appropriately so Tm's are compatible with whatever sequencing route you choose.

Submit at least 20 microliters at 100 micromolar concentration of each of your primers per each intended lane of sequencing.

== How long will it take for my HiSeq sample to be sequenced? ==
This is a very hard question to answer because it involves a number of variables. The simplest answer is, sample preparation typically takes ~1 weeks for standard samples with all reagents on hand (QC, prep, QC). For newer or rarely used methods, the number can extend to a month if we need to gather reagents. Once the sample is ready, it goes in to the sequencing queue. This queue is often the longest part of the process. Once the full flowcell is ready and put on a sequencer, it takes ~1.3h per base to sequence (this used to be 1h but upgrades to the machines have slowed them down some). A 40nt read takes just over 2 days. Turn around (for PE samples) functionally takes about a 1/2 day. So a 40+40 PE sample will take about 6 days once it is on the sequencer - if nothing goes wrong. Overall, when the queue is flowing at full speed, a 40nt SE submission with sample preparation can be returned with data in right around 3 weeks. 40+40 PE samples more typically take ~4-6 weeks for [[BioMicroCenter:CoreDeps|CORE lab members]] with the extra time coming from delays in filling out flowcells. 

NextSeq and MiSeq are much simpler. Once the sample is prepped and the machine is working, it will be loaded immediately. Total times should be less than 2 weeks.

=== Illumina Queue: Full HiSeq Flowcells === 
The primary requirement for a HiSeq run is a full flowcell. Each flowcell is composed of 7 lanes that must be run together. In the BioMicro Center, lanes are grouped by read length to optimize throughput and keep costs low. This requirement for 7 samples has a major impact on queue time. If you do a common read length, such as 40nt SE, the queue length will be short. On the other hand, if you ask for an unusual read length, your samples may never come off the queue. By frequency, the most common read lengths are, in order: '''40SE, 50SE, and 40+40PE'''. We do fudge a little in some of the long read lengths to fit samples in (mixing 40 and 50 for example), but we will always give priority to the samples of the read length we are going to do. We do not run incomplete flowcells and other options, such as NextSeq and MiSeq exist for unusual read lengths. NOTE: Barcoding is assumed to be done on all flowcells and we do NOT restrict pooling based on which samples are multiplexed and which are not.

=== Illumina Queue: Priority ===
{|
 |width=450|
The BioMicro Center queue operates on a tiered priority basis (see table). We hold to this priority schema as closely as we can while still needing to fill flowcells. Some caveats: Once a flowcell is quality controlled and "on deck" for clustering, that flowcell is considered locked and no new submissions will be added to it. Clustering can happen a week or more before loading. Once flowcells are clustered, their order is set barring the need for reruns.
 |
{| align="right", textalign="center", border=1 
 ! Priority
 ! Type
 |-
 | 1
 | Reruns of failures
 |-
 | 2
 | Samples from labs that bought the equipment 
 |-
 | 3
 | [[BioMicroCenter:CoreDeps|MIT CORE labs]]
 |-
 | 4
 | MIT non-CORE labs
 |-
 | 5
 | non-MIT users
|}
|}

=== Failures ===
The Illumina HiSeq is temperamental beast and failures do happen. The large majority of the failures on our sequencer are due to equipment failure and not due to operator failure, although those do happen as well. Once a failed run is identified, we must work with Illumina to identify the cause and prove to them that the cause is equipment based. This allows us to get replacement reagents which helps keep your costs down. Once a run has failed, the failed samples on that flowcell are given highest priority. 

Failures create a large challenge for us in providing estimates of when your data will be ready. Because the failure rate is so high and failures tend to cluster, short sequencing estimates can quickly turn in to long waits, particularly if you are not at the top of the priority queue. The BioMicro Center works closely with the KI Biopolymer Core and the Whitehead Institute GTC to move samples to other machines whenever possible if we develop a backlog. 

=== Triage ===
The final consideration in timing is best summarized as "Triage". This encompasses all of the many other factors we take in to account in creating flowcells. This can range from scheduling: Pushing a 40nt SE run on before an 40+40PE run so the turn around of the PE times to be during the week instead of over a weekend - to flowcell logistics: Grabbing low priority 40SE samples to fill out a rerun flowcell while an 40+40PE run that does not have rerun samples waits in the queue. Our goal is to get everyone their data as fast as we possibly can. We look at every project every day and we are working as hard as we can to get you your data as fast as we can.
-->