karafecho commented 2 years ago

Submitting Team

Multiomics EHR Risk Provider (Jenn H.)

SME

Philip Mease, MD, MACR, received his undergraduate and medical degrees from Stanford University and completed a residency in internal medicine and a fellowship in rheumatology at the University of Washington. He cares for patients at Seattle Rheumatology Associates, directs the Rheumatology Research Division at Swedish Medical Center/Providence St. Joseph Health, and is Clinical Professor of Medicine at the University of Washington in Seattle. Dr. Mease’s research interests include psoriatic arthritis (PsA) and spondyloarthritis (SpA), rheumatoid arthritis (RA), and osteoarthritis (OA). His seminal 2000 Lancet paper established the efficacy of anti-TNF therapy in PsA. He has authored over 500 journal articles and numerous book chapters and has edited textbooks on PsA and SpA. Dr. Mease is past president and founding organizer of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA), an international organization of over 800 rheumatology and dermatology researchers dedicated to research and education about psoriasis (PsO) and PsA. He is also an active member of the Assessment of Spondyloarthritis International Society (ASAS), the Spondyloarthritis Research and Treatment Network (SPARTAN), and the Outcome Measures in Rheumatology Clinical Trials (OMERACT) research organization. In addition, Dr. Mease is a Director of the Spondyloarthritis and Psoriatic Arthritis of the Consortium of Rheumatology Researchers of North America (CORRONA) and is co-chair of the PsA task force of the National Psoriasis Foundation, which in 2019 awarded him a lifetime achievement award for his work in advancing research on PsA.

Challenge Question

Simple question: What are risk factors for progressing from PsO to PsA?

Specific question: In a population of patients with PsO, what risk factors could we use to recruit a cohort of patients with increased risk for new onset of PsA within three years?

Background

The goals of this challenge question are to identify risk factors for new onset of PsA within three years among patients with PsO for the purpose of: (1) earlier recognition and treatment of PsA; and (2) recruitment of a cohort for a prospective multiomics study of pre-onset transition to PsA. In patients with PsO, known predictors of PsA include higher PsO severity, nail pitting, arthralgia, imaging-MSK inflammation, higher categories of BMI, family history, and select biomarkers. In addition, there are biomarkers associated with PsO severity and PsA severity. In outpatient-based cohort studies, the incidence rate of PsA per 100 patient-years varied from 1.34 to 17.4 (Zabotti et al 2021). For prospective multiomics studies of transition to PsA, one would like an enriched population where at 25% to 50% of patients experience a transition. Screening for the study could include medical/exposome history, physical examination, standard of care clinical laboratory measures, or novel biomarkers. However, ideally, it would not require WGS or other more expensive or invasive assays at initial time of recruitment. Rather, WGS and other multiomics assays would be conducted on biobanked samples for patients who transition to PsA during a prospective trial.

Please see https://pubmed.ncbi.nlm.nih.gov/34385474/ for additional background.

Please post GitHub queries and results to this ticket.

Additional Background (added August 5)

Transition period from PsO to PsA

roughly 10 years
symptoms progress slowly

Formal diagnosis of PsA

Caspar classification criteria
PsA vs non-PsA inflammatory disorder
evidence of inflammatory arthritis not osteoarthritis
psoriasis plus negative Rheumatiod Factor
psioriasis plus nail conditions, etc

Prodromal phase

achiness, instability
excessive fatigue
scalp psoriasis
psoriasis in folds of skin
toe nail disease
-omics markers suggested, e.g., TNF pathway, but tentative
often misdiagnosed as fibromyalgia

Timeline

August 5, 2022 - First Friday of the Month Standup

Check-in/update from teams on progress

August 12, 2022 - Second Friday of the Month Standup

Check-in/update from teams on progress

August 18, 2022 - Third Thursday of the Month Translator QotM mini-hackathon

Review answers and refine queries with SME; wrap-up challenge

August 26, 2022 - Fourth Friday of the Month

Translator QotM Challenge summary published in Gazette; next month's QotM Challenge announced

jh111 commented 2 years ago

Mondo Psoriasis (PsO): MONDO:0005083 Psoriatic arthritis (PsA): MONDO:0011849 psoriatic arthritis, susceptibility to, 1: MONDO:0100231 psoriatic arthritis, susceptibility to, 2: MONDO:0100232

Sample query

{
  "edges": {
    "e0": {
      "object": "n0",
      "subject": "n1"
    }
  },
  "nodes": {
    "n0": {
      "categories": [
        "biolink:PhenotypicFeature",
        "biolink:Disease"
      ],
      "ids": [
        "MONDO:0100231",
        "MONDO:0100232"
      ],
      "is_set": false
    },
    "n1": {
      "is_set": false
    }
  }
}

ARS result Older result

Bassa1 commented 2 years ago

PsA, MONDO:0011849, related to any disease:
Standard querying

{ "edges": { "e0": { "object": "n0", "predicates": [ "biolink:related_to" ], "subject": "n1" } }, "nodes": { "n0": { "categories": [ "biolink:Disease" ], "ids": [ "MONDO:0005083" ] }, "n1": { "categories": [ "biolink:Disease" ] } } }

ARS result

Bassa1 commented 2 years ago

*Standard Query { "edges": { "e0": { "object": "n0", "predicates": [ "biolink:developed_from" ], "subject": "n1" } }, "nodes": { "n0": { "categories": [ "biolink:Disease" ], "ids": [ "MONDO:0011849" ] }, "n1": { "categories": [ "biolink:Gene" ] } } } ARS Result

andrewsu commented 2 years ago

Regarding this line in the background:

In patients with PsO, known predictors of PsA include higher PsO severity, nail pitting, arthralgia, imaging-MSK inflammation, higher categories of BMI, family history, and select biomarkers.

Can we get a list of what those "select biomarkers" might be? I'm guessing they are all in that review article, but I suspect an SME would do a much better job of extracting/summarizing that information than a non-SME...

karafecho commented 2 years ago

MONDO:0005083 (PsO) PK = 880d7863-423c-4ed7-a032-f6c98cf7162c

MONDO:0011849 (PsA) PK = 77acf53d-d848-4984-917c-59eb59e4ff3c

{
  "edges": {
    "e0": {
      "object": "n1",
      "predicates": [
        "biolink:has_real_world_evidence_of_association_with"
      ],
      "subject": "n0"
    }
  },
  "nodes": {
    "n0": {
      "categories": [
        "biolink:Disease"
      ],
      "constraints": [],
      "ids": [
        "MONDO:0011849"
      ],
      "is_set": false
    },
    "n1": {
      "categories": [
        "biolink:DiseaseOrPhenotypicFeature"
      ],
      "constraints": [],
      "is_set": false
    }
  }
}

karafecho commented 2 years ago

@jh111 @Bassa1 : I'm wondering if you all are planning to use the new Biolink predicate?

associated with
     correlated with
          positively correlated with
          negatively correlated with
     associated with likelihood of

Bassa1 commented 2 years ago

@karafecho I am queering based on the new Biolink predicates

Bassa1 commented 2 years ago

@karafecho @jh111 I just chatted with Colleen about using the new Biolink predicates because I am not finding some of the predicates in the Translator/ARAX UI. She said the ARAX. UI may not be updated to the bio link 2.4.8 version. I sent a message on the UI slack channel

jh111 commented 2 years ago

New resources available in the QotM Psoriatic Arthritis directory, including the CURIEs for known risk factors, additional queries, the Psoriatic Arthritis Primer and CASPAR criteria, and tracking SME questions for Dr. Mease @andrewsu

marcdubybroad commented 2 years ago

However, ideally, it would not require WGS or other more expensive or invasive assays at initial time of recruitment.

Would it make sense to use translator to cheaply fill out the genetic association picture at the start? It's putting the cart before the horse, but if we:

do a two hop pso -> gene or pathway <- psa query
find the top gene/pathway hits
validate the data with the primer's expected genes/pathways (confidence in the translator results)
use any new gene/pathway results to find what other phenotypes they are associated with
look into those phenotypes as risk factors for pso leading to psa

jh111 commented 2 years ago

@marcdubybroad Great suggestion. For the first steps, in addition to the ones listed above, we could also try the following alternates.

PsO -> gene or pathway <- genes known to be associated with PsA genes (see list)
PsO -> gene or pathway <- medications used for PsA (but not PsO)

marcdubybroad commented 2 years ago

For the pso-pathway-psa query:

{
    "message": {
        "query_graph": {
            "edges": {
                "e0": {
                    "object": "pathway",
                    "subject": "pso"
                },
                "e1": {
                    "object": "psa",
                    "subject": "pathway"
                }
            },
            "nodes": {
                "pso": {
                    "categories": [
                        "biolink:PhenotypicFeature",
                        "biolink:Disease"
                    ],
                    "ids": [
                        "MONDO:0005083"
                    ],
                    "is_set": false
                },
                "psa": {
                    "categories": [
                        "biolink:PhenotypicFeature",
                        "biolink:Disease"
                    ],
                    "ids": [
                        "MONDO:0011849"
                    ],
                    "is_set": false
                },
                "pathway": {
                    "categories": [
                        "biolink:Pathway"
                    ],
                    "is_set": false
                }
            }
        }
    }
}

https://arax.ncats.io/?r=ae2b1e75-7eef-444f-a373-c199ff6e12c0 Note: we are in the process of loading about 300+ more pathway/phenotype associations (go, reactome, wikipathways, etc), although we don't have pso/psa, but the results from linked pathways above could be used to query those phenotypes.

Bassa1 commented 2 years ago

Thank you, Marc This is great! I queried with a single edge, but I haven't gotten results. Here is what I put in PsO increases_degradation_of pathway correlated_with PSA

karafecho commented 2 years ago

Queries from 08.18.2022:

PsO – Gene – PsA: https://arax.ncats.io/?r=8dddb496-127d-4fb8-951e-9047957d3fb3

PsA genes (7) – DiseaseOrPhenotypicFeature: https://arax.ncats.io/?r=a7ee2d96-2b73-4e01-b462-b978b19b0555

PsA genes (minus TNF) – DiseaseOrPhenotypicFeature: https://arax.ncats.io/?r=87bfe5ba-8d7c-4608-a04f-b3fcf4e0f3d6

PsA genes (minus TNF, is_set:true) – DiseaseOrPhenotypicFeature: https://arax.ncats.io/?r=609fe856-ae98-44a0-ba61-5b6dc325b90d

PsA MRO genes (is_set: true) – NamedThing: https://arax.ncats.io/?source=ARS&id=d2c0eea6-1a5a-4988-b90c-6049bfd82490

enthesitis (MONDO:0044999) – gene – gene – gene – scalp disorder (MONDO:0024419): https://arax.ncats.io/?r=1994cac1-f36e-47a0-b67a-5b672aca0dfd

enthesitis (MONDO:0044999) – gene – gene – gene – scalp disorder (MONDO:0024419): https://arax.ncats.io/?r=56948

enthesitis (MONDO:0044999) – NamedThing – NamedThing – NamedThing – scalp disorder (MONDO:0024419): https://arax.ncats.io/?r=56949

PsA genes (minus TNF, is_set:true) – DiseaseOrPhenotypicFeature: https://arax.ncats.io/?r=fe05df77-1f7e-4e1e-a422-26e1f429b787

PsA genes (minus TNF, is_set:true) – DiseaseOrPhenotypicFeature – AnatomicalStructure (connective tissue, UBERON:0002384): https://arax.ncats.io/?source=ARS&id=6fa04129-c12e-4c47-9975-8db33c81796a

PsA genes (minus TNF, is_set:true) – DiseaseOrPhenotypicFeature – NamedThing (connective tissue, UBERON:0002384): https://arax.ncats.io/?r=4b834c60-29cd-452a-aa0a-ec9c3cc37dc3

type 2 diabetes – Gene – PsA: https://arax.ncats.io/?r=457d390d-df3a-4d45-81e3-d3aa0bd8fbb6

Bassa1 commented 2 years ago

@karafecho Thank you, Kara, for these queries. Here are some additional queries we have been working on: Anything->related/subclass_of/-> psoriatic arthritis susceptibility : https://arax.ncats.io/?r=c823f15f-8857-4611-83f9-b968a895ac2a DiseaseorPhenotypicFeature related ->has_real_world_evidence_of_association_with-> PsA: https://arax.ncats.io/?r=8527c883-37d5-41d2-b22a-b46893e1b9d3 Proteins ->associated_with-> PsA: https://arax.ncats.io/?r=06f75fed-8584-4780-b569-2934daec62ce Genes ->associated_with->PsA: https://arax.ncats.io/?r=80f6f6af-4e90-4dd6-a45a-472a1e03f723 Disease ->has_real_world_evidence_of_association_with-> Protiens->has_real_world_evidence_of_association_with -> PsA:https://arax.ncats.io/?r=1043bf91-d6ca-4e34-900e-e330902b3022

jh111 commented 2 years ago

ARS query for Anything associated_with_increased_risk_for psoriatic arthritis

karafecho commented 1 year ago

Closing with comment in #233 ...

NCATSTranslator / testing

Question of the Month #5: Psoriatic Arthritis #220

Submitting Team

SME

Challenge Question

Background

Additional Background (added August 5)

Timeline