Scribd
Upload
2 views

Workflow

The document outlines a new streamlined workflow for chip quality control (QC) and data processing, emphasizing rapid data curation and diagnostic capabilities using GPU technology. It details migration considerations for data pipelines and the integration of various applications, along with proposed timelines for implementing the Auto-Correlation Function (ACF) and Matrix Profiling Analysis (MPA) into the Pipeline Management Repository (PLMR). Key milestones are set for validation, testing, and deployment phases throughout the latter half of the year.

Uploaded by

Paul Mudondo
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2 views

Workflow

The document outlines a new streamlined workflow for chip quality control (QC) and data processing, emphasizing rapid data curation and diagnostic capabilities using GPU technology. It details migration considerations for data pipelines and the integration of various applications, along with proposed timelines for implementing the Auto-Correlation Function (ACF) and Matrix Profiling Analysis (MPA) into the Pipeline Management Repository (PLMR). Key milestones are set for validation, testing, and deployment phases throughout the latter half of the year.

Uploaded by

Paul Mudondo
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 3

New Streamlined Future Proofed

Workflow
PC

Initial Chip QC report and ACF curation report available in < 1 hour to
H5 File
inform next expt. And accelerate cycle of learning MPA Report visualization

PC deployed apps

Reports:
Reports: Chip QC, DBA, Wizyme, HMM,
DL

TDMS File Chip QC HMM


Raw System (Python) H5 File Local
Experiment Data Retention Server
Data Chunking MPA
ACF deployed
DL apps
Metrics & report automation Chip Health
Simultaneous GPU Based (MatLab)
Data Curation & Chip Diagnostic
Server Cloud DL
GPU Leveraged Gen3 Local Workstation Backup REMS LIMS
deployed
Bottleneck eliminated by rapid local data processing prior to apps HMM
transfer to local server.

Local server and cloud


CONFIDENTIAL - Property of Roswell Biotechnologies Inc.
Migration considerations
•Data pipelines/processes that ingest data from various data sources
•The data stored in the existing databases/storage (s3).
•Scripts and procedures to manipulate and further process and analyze the data.
• Do we need to rewrite our data pipelines? This needs to work in a Hybrid
environment (cloud + on-premise)
•A business application that reads or interacts with the data.

Data provider process - BigQuery - end client application (JMP/Analysis etc – Looker
integration)

Schema design – partitions, clusters  repeated and nested fields from flattened tables

CONFIDENTIAL - Property of Roswell Biotechnologies Inc. 2


PLMR: PipeLine Management Repository
Proposed Timelines for ACF-MPA/HMM PLMR ACF: Auto-Correlation Function
MPA: Matrix profiling analysis
HMM: Hidden Markov Model

July August September October November December

ACF 4.2 Validation Milestone:


Chip activity & ACF

Stable V1
QC/DBA/Wizyme routines Lock Down
ACF to PLMR
Testing on Stations

Artemis I - PLMR, Chip Health, ACF


Fundamental Chip Metrics R&D
Chip Metrics

PLMR alpha
ACF -> H5 on PLMR

ACF->H5->MPA on PLMR

ACF->H5->HMM on PLMR
QC & PLMR beta
ACF, PLMR QC & Matlab in parallel PLMR QC, ACF

Aug 28th Sep 30th Nov 25th


Milestone: First experiment Milestone: Milestone:
with local processing. Deploy to all users End of MatLab chip health
CONFIDENTIAL - Property of Roswell Biotechnologies Inc.

You might also like